[00:13] nvidia compatible with natty is in ppa:ubuntu-x-swat/x-updates now (and this time I made sure it actually worked) [00:13] 6 hour build queue though :( [00:15] Oh, funky. That's released? [00:16] Sarvatt: \o/ [00:16] yay!! [00:17] Good news, I sware my Apple notebook has almost been cooked from the inside out whilst running nouveau in 2D only. :) I know it will cook with 3D. :) [00:18] 7;3~/c [00:21] RAOF, released and public :-) [00:22] * Laney looks forward to minecraft again :-) [00:24] Laney, so much for your productivity eh? [00:25] Sarvatt: whoo! === Bertrand is now known as bl8 [01:35] bug 724678 is the best bug report, ever [01:35] Launchpad bug 724678 in nautilus "upgrading" [Undecided,New] https://launchpad.net/bugs/724678 [01:39] wow. that's impressively useless === james_w` is now known as james_w === asac_ is now known as asac [03:05] looking to change my maverick netbook install desktop back to the the one found in lucid [03:05] but I cant figure out how to do that - any hints, suggestions ? [03:10] jwhisnant: It's not clear exactly what you want to do. However, #ubuntu-desktop is unlikely to be the correct forum for answers; this place is for development. [03:10] !support | jwhisnant [03:10] jwhisnant: The official ubuntu support channel is #ubuntu. Also see http://ubuntu.com/support and http://ubuntuforums.org and http://askubuntu.com [03:11] RAOF, thanks, I'll ask there ... again ... [03:11] If #ubuntu isn't helping, then ubuntuforums or askubuntu, although slower, should have a broader audience. === Bertrand is now known as bl8 === Bertrand is now known as bl8 [03:59] w00t, uploaded the last firefox beta to natty now \o/ [04:00] chrisccoulson: Hurray! So that should be built, what, early next week? :) [04:00] lol [04:00] it doesn't take quite that long ;) [04:00] it's about 2.5 hours, and most of that is the test suite ;) [04:01] (all but 40 minutes) [04:05] * micahg hopes 2 of the remaining build failues disppear [04:19] bryce_, I am in X dependency hell :\ [04:23] lamalex, oh? [04:23] raof or tjaalton or I could probably help but we'd need to know more specifics [04:23] bryce_, yeah I need to know what more specifics to give you before I can give them :P [04:23] Well, let's start with the error message :) [04:24] let's move to #distro? [05:56] Morning all! [05:58] hi Sweetshark [05:59] * Sweetshark suspects himself to have put a buildd in an endless loop ... [05:59] *ashamed* [07:54] good morning [07:55] waow, 15 minutes blaming first my wifi spot before seeing that dhcp3-client was the cause of not connecting… [07:57] ok, seems that isc-dhcp (4.1.1-P1-15ubuntu5) fixes it [07:57] just upgraded at the wrong moment it seems :) [08:01] Good morning [08:01] hey pitti! How are you? [08:02] hey didrocks [08:02] for some reason terribly tired [08:02] but I'll manage [08:02] oh? the week-end will be there soon :) [08:02] me too! [08:02] hey mvo ;) [08:02] * mvo yawns and takes a extra ccup of tea [08:03] hey didrocks [08:03] * pitti hugs mvo [08:03] * mvo hugs pitti [08:04] didrocks: FYI, http://launchpadlibrarian.net/65082307/buildlog_ubuntu-natty-amd64.folks_0.3.6-0ubuntu1_FAILEDTOBUILD.txt.gz used the broken dbus [08:04] didrocks: I uploaded a fix an hour after 1.4.6-1ubuntu1 when I noticed the same issue with avahi [08:04] it can just be given back, doing now [08:05] pitti: thanks for the notice :) [08:05] so, where did my window decorations go.. [08:05] pitti: is unity-window-decorator running? [08:05] pitti: have you just upgraded? [08:05] partial upgrde last night [08:06] doing dist-upgrade now [08:06] did you restart since? [08:06] didrocks: no u-w-d [08:06] didrocks: yes, booted this morning [08:06] there is an ABI break in compiz, u-w-d has to be restarted after the upgrade [08:06] I'll try a full dist-upgrade first [08:06] hum… [08:06] ok :) [08:06] it always takes me a bit to sort out [08:06] X.org has been broken for many days now [08:06] I'm testing upgrading on my netbook now just to check [08:06] trying to dist-upgrade everything removes ubuntu-desktop [08:07] Xorg is still hang because of nvidia here :/ [08:07] hum… [08:07] so, the transition yesterday was quite horrible (I had to update and not just rebuild) 17 packages [08:07] was some package renaming and such [08:07] * pitti hugs didrocks, great job! [08:08] let me see on my intel box if it's ok [08:08] * didrocks hugs pitti back, I hope it's not because of me that ubuntu-desktop wants to be removed :) [08:08] let's see on my (slow) netbook [08:11] ok, the places with the new libunity symbols and dee are installed, it kept back unity (and all compiz friends) though [08:11] ah libindicator transition :) [08:14] hm, dist-upgrade is still horribly broken [08:15] yeah, confirmed, killall unity-window-decorator && unity just after the upgrade works [08:15] (removed libindicator2 for forcing the upgrade) [08:15] good morning guys [08:16] hey kenvandine, is it you who told me yesterday evening that it was late, it's time to go to bed, blablabla… ? :-) [08:16] hehe [08:16] soon :) [08:16] didrocks: upgrade fails because compiz-plugin-mains file conflict with compiz-fusion-plugins-main [08:16] pitti: argh, the Replaces: [08:16] didrocks: it seems the former doesn't have a strict enough Conflicts: ? [08:17] pitti: yeah, transitional package, stupid issue from me, fixing [08:17] hm, it doesn't have a C:/R: at all [08:17] didrocks: thanks [08:17] * bryce_ waves [08:17] pitti: yeah, I was interruped when I was doing it and forgot then :/ [08:17] didrocks: np; I have to install pacakges one by one, great way to find those :) [08:17] lol [08:17] hey bryce_ [08:17] * mvo hugs kenvandine [08:17] pitti: the compiz transition was quite… epic (upstream rolled 6 times the tarball before having something which works) [08:17] heya pitti [08:18] hey mvo [08:18] didrocks, ugy [08:18] hey :) [08:18] ugh even [08:18] kenvandine: and imagine I compiled all buggy version :/ [08:18] (or tried to…) [08:18] so why does upgrading libunity3 remove empathy.. [08:18] funny story [08:18] pitti, amd64? [08:18] yes [08:19] empathy is dep wait [08:19] now that we FINALLY have binary drivers that work with the latest xserver ABI, guess what [08:19] pitti: empathy -> folks [08:19] bryce_, this isn't going to be funny is it? [08:19] empathy Depends: libunity3 (< 3.4.4) [08:19] yep, X.org has changed the ABI in xserver again [08:19] bryce_: argh! [08:19] pitti, folks had failed to build on amd64 earlier, the dbus pc file thing [08:19] kenvandine: right, retried [08:20] pitti, so i just rebuilt that [08:20] kenvandine: sorry about that, we had a broken dbus in the aarchive for just one hour, and it must have caught that [08:20] waiting for it to publish so empathy can build [08:20] yeah, no worries :) [08:20] kenvandine: out of interest, why would empathy have a < dependency on libunity? [08:20] that was libunity's fault [08:20] shlib [08:21] that changed in today's libunity upload though [08:21] right didrocks? [08:21] kenvandine: ah, typo in the shlibs then? [08:22] didrocks, can you make sure empathy gets built after folks for amd64 gets published? i am exhausted... [08:22] file replacement problem in python-aptdaemon, I'll fix that myself [08:22] not a typo, i think it as intentional [08:22] kenvandine: how can this be intentional? [08:22] kenvandine: anyway, we'll handhold empathy & co, you should go to sleep [08:22] the shlib was >= 3.4.2 and << 3.4.4 [08:23] something like that [08:23] kenvandine: ok, that disqualifies as a typo, but it's still wrong [08:23] it's been a long day :) [08:23] pitti: it wasn't a typo, let me explain you :) [08:24] but first, uploading the fix C/R [08:24] pitti, didrocks had reasons... but it is history now :) [08:24] anyway... thanks guys [08:24] time for some sleep [08:25] good night kenvandine [08:25] and good morning to you :) [08:25] * kenvandine waves! [08:25] kenvandine: sleep well! [08:27] pitti: ok, done. sorry about it :( [08:27] didrocks: no problem, thanks for quick fix! [08:27] I thought about the transition, thought about debian syncing, but got interrupted when doing the transitional package [08:27] :/ [08:28] pitti: so, on libunity [08:28] aptdaemon collected a fair number of upgrade bugs already [08:28] we decided with the dx team that the libunity API/ABI was totally unstable until we have "real" consumers and feature freeze is there [08:28] (unity-places* aren't considered as real consumers) [08:29] so, they broke a lot ABI/API recently [08:29] didrocks: ah, I see [08:29] that's why the packaging was >= and << [08:29] didrocks: so this is just a hack until the final release? [08:29] right :) [08:29] since yesterday, it has now proper symbol tracking [08:30] (note that nux still >= and << as the ABI won't be stable soon, but unity is the only consumer) [08:30] * pitti uploads aptdaemon fix [08:32] pitti: There is a 3.3.1-1ubunutu1 in the ppa with human icons and it actually compiled in finite time. \o/ [08:33] bryce_: is it package already or it won't do it? [08:33] pitti: Do you have a minute to discuss the libreoffice 3.3.1 backporting to maverick mess? [08:35] Sweetshark: niiice! [08:37] bryce_, RAOF: I have had a broken x.org upgrade (package removals) for a few days now; is that known? [08:37] xserver-xorg-input-evdev : Depends: xorg-input-abi-12.1 [08:37] xserver-xorg-input-mouse : Depends: xorg-input-abi-12.1 [08:37] xserver-xorg-input-vmmouse : Depends: xorg-input-abi-12.1 [08:37] xserver-xorg-video-vmware : Depends: xorg-video-abi-9.0 [08:37] I had a long at the 30 hours buildlog and found nothing too suspicious. Seems the builder was just too slow for that tiny package LO [08:38] Sweetshark: I hope the 3.3.1-1ubunutu1 was just an IRC typo, it's not in debian/changelog? [08:38] ubun"u"tu, I mean [08:38] Sweetshark: maverick backport> you mean in a PPA? [08:40] pitti: do you have on your keyboard some umlaut caracters directly accessible? [08:40] like ö [08:40] or do you have to press two keys for that [08:40] didrocks: I normally use US layout, but have German layout configured as well; if I need them, I just switch [08:41] didrocks: in German you have ä, ü, ß, etc. [08:41] pitti: yep, just a typo [08:41] pitti: and for them, you have a direct key for that (no need for " + o for instance)? [08:42] didrocks: yes [08:42] pitti: ok, thanks :) [08:42] didrocks: want to learn German? [08:43] pitti: well, I've done 6 years of German, unfortunately, not a lot of things was kept in my memory :) [08:43] pitti: I was figuring if bug #724747 was valid [08:43] Launchpad bug 724747 in unity "Super key shortcut overlay not working with umlauts, greek/cyrillic letters" [Undecided,New] https://launchpad.net/bugs/724747 [08:43] (because a shortcut need a direct key, no key combination :)) [08:43] pitti: sure your mirror is just stale? here evdev depends on -input-abi-12 which is correct [08:44] 2:1.9.99.902-2ubuntu1 0 [08:44] 500 http://archive.ubuntu.com/ubuntu/ natty/main amd64 Packages [08:44] *** 2:1.9.99.901+git20110131.be3be758-0ubuntu6 0 [08:44] 100 /var/lib/dpkg/status [08:44] ooh [08:44] that must be a leftover from the time when I was using the PPA [08:44] pitti: yes, maverick backport in the ppa. Our glorious debian/rules generates the debian/control from debian/control.in. And it does it different for maverick and natty. Thus, my debian tarball for 3.3.1-1ubuntu1 maverick was different from the one for natty, which made dput very angry after the upload. [08:45] Sweetshark: you need to change the version number anyway [08:45] Sweetshark: 3.3.1-1ubuntu1~maverick [08:46] Sweetshark: even if the source package was absolutely identical (which it can't, because changelog has to say natty vs. maverick), the binaries would build against older libraries [08:46] dpm: good morning - i registered lp:giraffe if you want to propose a merge [08:46] and thus the .debs wouldn't be identical [08:46] hey kamstrup, good morning, cool, thanks :) [08:46] dpm: and I also added you to the giraffe-dev team so you have commit rights [08:46] bryce_, RAOF: unping then, local failure; got it sorted out now [08:47] kamstrup, thanks [08:48] dpm: just whipping up an answer to your mail [08:49] great [08:49] Sweetshark: oh, we need a corresponding -l10n package upload as well, don't we? or does the current 1:3.3.0~rc4-1ubuntu3 work with 3.3.1? [08:50] Sweetshark: I'm testing an upgrade to the PPA now to check for the -l10n compat [08:50] pitti: yep, I will need to do a l10n upload for natty. [08:50] hey everyone, what's the best package to file a bug about the notification popups? [08:52] dpm: notify-osd [08:52] although l10n _should_ not have changed from 3.3.0 to 3.3.1, but it might contain fixes. [08:52] thanks pitti [08:57] Sweetshark: it seems to work well with the current -l10n [08:58] Sweetshark: so, I'm happy to sponsor this to natty now, all ok from your side? [08:58] yep [08:58] what about FF? [09:00] Sweetshark: are there so many new features between 3.3.0 and 3.3.1? [09:00] in the past these were bug fixes (mostly anyway) [09:00] Sweetshark: but anyway, we wouldn't want to release natty with a release candidate when there's a newer final bug fix release out [09:02] Sweetshark: I'll spend a minute trying to craft a better source.changes, so that you see all changes since natty, not just the single "merge remaining changes from debian up to 3.3.1-1 [09:02] bah, empathy can't be installed on natty :( [09:02] hello pitti [09:02] Sweetshark: FYI, you can do that more easily during source package build, next time you do an upload I'll tell you [09:02] hey seb128 [09:02] xclaesse: on amd64? being worked on [09:02] yeah, amd64 [09:03] xclaesse: should be sorted out in about two hours [09:03] dpm: you've got mail :-) [09:03] pitti, great :D [09:03] thx [09:03] pitti: k, will do. [09:03] pitti, how are you? [09:05] seb128: bit tired, but ok otherwise; did you sleep better last night? [09:05] 3.3.1 is really a bugfix release, so it is ok I guess although with LO there is still a lot of change even on the release branch. [09:05] pitti, yeah, I slept well this night and feel better today, thanks [09:05] Sweetshark: oh, we need to sync openoffice.org from Debian for that, right? [09:05] seb128: great! [09:06] dput -u libreoffice_3.3.1-1ubuntu1_source.changes [09:06] wheeee! [09:07] Sweetshark: congratulations! that's your first Ubuntu upload! [09:07] * pitti hugs Sweetshark [09:07] pitti: we might even need our own "release" for the openoffice.org transitionals. [09:08] Sweetshark: ah, so we do need Ubuntu changes? [09:08] Sweetshark: should I sync the package for now, or do you want to prepare a custom upload right away? [09:08] * Sweetshark jumps around happily and vainly takes a peek at his karma in launchpad. [09:08] Sweetshark: karma will take a bit to catch the upload [09:09] * Sweetshark wont get tired of hitting F5 [09:09] hehe [09:09] Sweetshark, congrats! [09:10] kamstrup, got it, reading... :-) [09:10] pitti: I would think I take a look the openoffice.org package. We already have the 3.3.0 transitionals in natty, they should do for now, right? [09:11] Sweetshark: I thought you dropped them from the current libo? [09:11] morning [09:11] so no need to sync openoffice.org from debian right now. [09:11] didrocks: ah, window decorations back to normal after full upgrade [09:12] Sweetshark: the old transitional packages from the previous version will be NBS, so it'll do for now [09:12] Sweetshark: oh, "NBS" -- means "not built from source" [09:12] Sweetshark: i. e. the .debs stay in the archive until we clean them up manually, but they do not have any source package which builds them [09:12] Sweetshark: we can't release with NBS packages, but for a limited time period they are ok [09:12] pitti: nice :) [09:12] pitti: understood. [09:13] Sweetshark: can you please also prepare a matching -l10n? [09:13] * pitti disappears for breakfast/shower for a bit [09:15] pitti: will do [09:19] didrocks, well won't do it for now. It'd be fairly sadistic to break -nvidia now that we finally have a working driver [09:20] bryce_: ok :) [09:21] I think we'll have to digest the abi change at some point though. But not right now. [09:21] bryce_: oh btw, did you rememberabout my magnetic effect issue? [09:21] bryce_: it seems I only have it when I have just one monitor connected [09:21] bryce_: when I'm using twinview with two monitors, no issue… [09:21] didrocks, I haven't forgotten, sorry it's still bugging you [09:22] now that's plain old weird [09:22] yeah [09:22] I think we should wait for the next nvidia driver anyway [09:22] didrocks, ok works for me [09:23] you've been the only person noticing this bug so far, so might be something quirky with your hw [09:23] bryce_: loicm as it as well [09:23] has* [09:24] and same for him, only with unity enabled [09:24] didrocks, hm that makes it sound worse [09:25] right [09:25] didrocks, I think the next action for the bug is probably to forward it upstream [09:25] bryce_: yeah, but they will ask to test with the latest nvidia first :-) [09:25] as it sounds like we're out of ideas on things to try here [09:25] didrocks, yep [09:25] so, that's the pre-next action ;-) [09:26] heh, right :-) [09:26] meanwhile, you can play with wayland :-) [09:27] bryce_: hum, I need a working unity you know :-) [09:27] nice work on wayland btw! [09:29] didrocks, I hope in natty+1 we maybe can get unity+nouveau to be not such a massive headache [09:30] nouveau needs a lot of TLC, but maybe we can do it [09:31] bryce_: oh? you think it's achievable? [09:32] didrocks, I think it's a worthy goal for us X guys [09:33] that would be excellent :) [09:33] didrocks, we need to be able to provide you with a FOSS driver that doesn't suck [09:34] anyway, nice chatting, cya later! [09:34] bryce_: yeah, right now, there are a lot of mem leaks on nouveau if I understood correctly [09:34] bryce_: have a nice evening :) [09:43] Sweetshark: btw, you need to close the human icons bug manually [09:43] (will give you more karma, yummy!) [09:44] Sweetshark: the usual way to do that is to write a line in the changelog like [09:44] * Add back Human icons. (LP: #123456) [09:44] Sweetshark: the "LP: #xxxx" syntax is magic, and will cause LP to close the bug with the changelog on upload [09:46] pitti: k [09:46] Sweetshark: oh, I'll change the seeds back from libreoffice-style-tango to libreoffice-style-human, ok? [09:47] Sweetshark: (do you know "seeds"?) [09:48] no idea [09:48] The new shortcuts in the unity launcher are awesome! However, I've got a question: the old Super+W and Super+A shortcuts, have now been replaced to show all workspaces and the applications place, respectively. Does anyone know if there are new shortcuts for the old functionality (i.e. Super+w to show all windows in the current workspace, Super+a to show windows in all workspaces)? [09:48] Sweetshark: so, the "seeds" are a list of top-level packages which we want in a default install [09:48] just guessing... just guessed right. [09:48] Sweetshark: a program "germinate" computes the full set of pacakges on the CDs from that, including the transitive dependencies [09:48] Sweetshark: i. e. it says which apps/pacakges we actually "want" [09:49] Sweetshark: we "want" empathy, we don't "want" libc6 or libreoffice-common [09:49] pitti: does that include "suggests"? [09:49] Sweetshark: no, only depends and recommends [09:49] by definition, those ^ are installed by default, suggests aren't [09:49] Sweetshark: http://wiki.ubuntu.com/SeedManagement FYI [09:50] Sweetshark: oh argh, libo doesn't build -- it's waiting on an unavailable build dependency (this is called "depwait") [09:50] https://launchpad.net/ubuntu/+source/libreoffice/1:3.3.1-1ubuntu1/+buildjob/2285894 [09:50] ah, ok. Because the libreoffice-gtk integration suggests tango, which still makes IMHO ... [09:50] Sweetshark: apparently it grew a new build dependency to "epm", which is in universe [09:50] meh [09:50] Sweetshark: do we really need epm? [09:51] I just wondered myself. [09:51] Sweetshark: in this case there's two options: (1) drop the dependency, (2) move the package to main, which requires a thorough review, and will only be done for packages which we can sanely support [09:51] we shouldnt, but maybe we do because of the way lo-build/debian does this. [09:52] Sweetshark: is it actually called during package build, or do you only need it to prepare the source pacakge? [09:58] pitti: meh, thats another tricky one. We need epm for the smoketest. And there is code that disables smoketest on buildd builds: that is, it the control file is build on a buildd build, the var ON_BUILDD is set so that rules would generate a control file without epm and without smoketest. [09:58] can someone review/merge/upload this -> https://code.launchpad.net/~rodrigo-moya/ubuntu/natty/gnome-settings-daemon/fix-666806/+merge/51177 please? [09:59] Sweetshark: hang on, it regenerates debian/control on the buildds? uh [09:59] pitti: I wonder how that should work, generating the control file on buildds. [09:59] Sweetshark: if we won't need it anyway then, perhaps the simplest solution is to just drop it [09:59] Sweetshark: it generally doesn't; it's a very bad practice [10:00] pitti: right, thats strange. [10:00] rodrigo_: will do [10:00] pitti, can you upgrade ppa builds score? [10:00] pitti, thanks, it's from the patch pilot thing [10:01] pitti, or does score only work for the main builders? [10:01] seb128: yes [10:01] seb128: and no [10:01] pitti, can you raise the score on https://edge.launchpad.net/~ubuntu-desktop/+archive/ppa/+buildjob/2285941 ? [10:01] pitti, compiz's update crash on start here [10:01] I could just regenerate the control file with "./debian/rules control ON_BUILDD=y", edit the changelog to ubuntu2 and reupload ... [10:01] seb128: done [10:01] pitti, I've uploaded a debug version to the ppa [10:01] pitti, thanks [10:02] Sweetshark: heh, sure [10:02] Sweetshark: btw, you built the previous package with '-sa', right? [10:02] Sweetshark: as we now already have the orig tarballs in Launchpad (and in the PPA), you don't need -sa any more; will be a much smaller/quicker upload then [10:03] i still have not got my head around all the switches. I guess I did a debuild -S. [10:03] Sweetshark: bug 696527 should be "fix released" [10:03] Launchpad bug 696527 in libreoffice "LibreOffice - Human icons theme disabled, patch needs an update" [High,Fix committed] https://launchpad.net/bugs/696527 [10:04] Sweetshark: ok, so debuild -S was clever enough to include the orig then; nevermind [10:04] Sweetshark: "fix committed" is the right state when you put a fix into some git or bzr, but haven't uplaoded yet [10:04] Sweetshark: (sorry about all those new things today..) [10:05] pitti: well, that is exactly the state currently: fixed, but not uploaded [10:05] Sweetshark: hmkay, due to the current "doesn't build" [10:06] pitti: or could I upload a fixed 1ubuntu1? Probably not. [10:07] Sweetshark: no [10:07] once the source is in LP, it is set in stone [10:07] k [10:07] Sweetshark: fortunately there is a sufficient amount of natural numbers :) [10:16] hello everyone! the gnome-terminal can only be resized in fixed-sized steps, thus it's not working well together with the grid plugin. is this something worth a bug report or rather something we have to live with? [10:18] pitti: should use debuild -S again, or is there something better suited? [10:20] Sweetshark: no, that's _the_ command to build a source package (unless you use git-buildpackage and friends) [10:20] htorque: here I can resize it in the steps of the size of a letter -- that seems right to me? you certainly don't want to see half a line somewhere [10:22] pitti: should upload to the ppa again? im not allowed to main? [10:23] pitti: right, it makes sense, it just looks odd when eg. splitting two gnome-terminals to each half of the desktop (you'll get empty space in between) [10:23] Sweetshark: I think the quickest would be if you could scp debian.tar, dsc, and source.changes to chinstrap [10:24] pitti: if you maximize a gnome-terminal, you don't see half lines at the bottom [10:24] Sweetshark: I'll sign the changes there, and upload [10:24] htorque: ah, I never actually wondered how it does that; perhaps just coincidence that it usually looks like it'd just fit [10:28] here's what i'm talking about: http://img.xrmb2.net/images/841674.png - it's not the end of the world, but it's "natty-like" if other windows do completely extent to the borders :) [10:29] *not "natty-like" [10:31] pitti: files are in my ${HOME} on chinstrap, where should I move them? [10:32] Sweetshark: that's fine [10:32] doh [10:32] why can't intel write a driver which doesn't crash when you vt switch [10:32] seb128: you missed the debian.tar.gz [10:32] ;-) [10:32] sorry [10:32] pitti, what? [10:32] Sweetshark: you missed the debian.tar.gz [10:33] oh ok [10:33] Sweetshark: (and the source.build isn't necessary) [10:34] TheMuso: can you please have a look at bug 681412? I've got a question for you there; thanks! [10:34] Launchpad bug 681412 in at-spi "Can not enter password for Administrative tasks using Onboard Keyboard" [Medium,Triaged] https://launchpad.net/bugs/681412 [10:34] pitti: now its there [10:36] Sweetshark: sent launchpadwards [10:36] Sweetshark: thanks! [10:40] dbarth, didrocks: do you know whether https://launchpad.net/ubuntu/+spec/packageselection-dx-n-unity-a11y is up to date? it seems quite behind [10:41] pitti: let me look [10:41] I would say atk isn't finished, so all are in progress as it should [10:41] (well, I mean, reflecting the reality of the code) [10:42] so should we move that to beta-1 then? [10:42] a3 freeze is next Tuesday, seems unlikely to land before that? [10:43] pitti: still some atk thing will land on monday, so maybe waiting for that? [10:43] ok [10:43] btw, would be nice if someone can tackle bug #723782 [10:43] Launchpad bug 723782 in gnome-session "Finalize gnome-session fallback detection" [Undecided,New] https://launchpad.net/bugs/723782 [10:44] I'm trying to clean up http://people.canonical.com/~platform/workitems/natty/canonical-desktop-team-natty-alpha-3.html as that is despeartely behind [10:44] I tried to describe everything there [10:44] as I certainly won't have the time to do it to focus on unity hacking [10:45] didrocks: ok, will have a look at find someone [10:46] pitti: thanks a lot! [10:46] if it needs more explanation, I can give them [10:51] pitti: i'm updating the bp at the moment, this one is not up to date thankfully [10:51] dbarth: *phew* :) [10:51] dbarth: the branches didn't land [10:51] just some part of them did [10:53] didrocks: will Alt+F2 land for a3, or should we move to b1? [10:53] pitti: I moved it already [10:54] pitti: I'll start working on it after alpha3. I'm tackling today and Monday unity bugs for alpha3 [10:54] pitti: I've refreshed quite a few WI [10:54] ah, cheers [10:55] alt + F2 will be tricky though, as I have to learn vala very quickly and how to use dee :) [10:56] didrocks: if the homescreen works well, alt-f2 should be mapped to it instead [10:56] dbarth: for alpha3 as a workaround? it can [10:57] dbarth: not for the definitive solution then [10:57] Sweetshark: https://launchpad.net/ubuntu/+source/libreoffice/1:3.3.1-1ubuntu2/+buildjob/2285989 :/ [10:57] pitti, you got email [10:57] Sweetshark: seems this got qutie a lot of new dependencies :/ [10:57] Sweetshark: to avoid more upload/fix cycles, coudl you please install ubuntu-dev-tools, and run "check-mir" once? and then pastebin the result? [10:58] Sweetshark: run in the unpacked libo source tree [10:58] pitti: meh, I know my mistake [10:59] seb128: cheers; so I guess we postone [10:59] pitti, it's really a shame [10:59] Sweetshark: oh, lucene isn't necessary? or do we need a mir? [10:59] pitti, not sure what we have now is better security wise [10:59] it's just that it got in before there was security reviews [11:00] might be worth talkign to kees about it; if it's no worse, then this shoudln't be a blocker [11:00] so we have a non maintaining buggy choice instead of a working one just because the new one has issues [11:00] pitti: I am rebooted to maverick as I did want to go deaf by a nonthrottled nvidia fan and didnt DISTREL=natty when creating the control file :/ [11:00] but the old one is not any better... [11:00] pitti, right, will do that when he's online later today [11:00] seb128: ok, cheers [11:01] Sweetshark: heh; so, take III [11:01] Sweetshark: just to be sure, after you regenerated control, could you please still run check-mir? [11:02] good morning everyone [11:02] hey chrisccoulson [11:02] didrocks, you didn't see that i pushed a commit to compiz-plugins-main ;) [11:02] i was going to just upload it until i realised that i couldn't [11:02] chrisccoulson: oh? you uploaded it? [11:02] chrisccoulson: oh bad… [11:03] chrisccoulson: weird, bzr didn't complain on bzr pull and bzr push… [11:03] hmmm :/ [11:03] chrisccoulson: sorry for that, it would have been cool if you could have uploaded that! [11:03] chrisccoulson: you need to apply for core-dev :) [11:03] heh :) [11:09] dpm: hey, how are yuou? [11:09] dpm: OOI, are there any news wrt. "Talk to Mozilla people about using user provided translations from Launchpad"? [11:11] pitti, doing well, thanks. I'll see if I can get to it today. I wrote the e-mail a while ago, and sent it for review to chriscoulson and rickspencer3. I just need to rewrite it to factor in the recommendations from Rick re: the Mozilla agreement [11:12] ah, thanks [11:16] pitti: I still get epm as dep. [11:16] Sweetshark: hm, I thought that got dropped with ON_BUILDD=y? [11:19] there are several natty packages that need a rebuild with the new GTK3 package names, is anyone working on those? [11:19] pitti: the if is a bit deeper nested, so I got it wrong. If we have java enabled (we do) and have no openjdk (but we have it), the smoketest would be disabled on buildd, but with openjdk we have the smoketest enabled still. [11:20] rodrigo_, which ones? [11:20] seb128, let me pastebin [11:21] Sweetshark: would it be adequate to disable the smoketests on the buildds regardless? [11:21] seb128, http://pastebin.com/qTuMHhuc <- some of them are from the gnome3 ppa, so ignore those [11:22] pitti: not really, as you dont really want a buildd and non-buildd build to have different deps. But I could disable the smoketest for natty by default for now. [11:22] but gnome-applets, indicator*, etc [11:23] Sweetshark: or that [11:23] seb128, I'm preparing a branch for them, if nobody is working on it [11:23] Sweetshark: coudl it be changed to "if epm is installed, run the smoketest"? [11:24] rodrigo_, not sure what your issue is coming from [11:24] oh wait, gnome-applets doesn't need it [11:24] rodrigo_, I just checked on my natty system, only the new libnotify and mousetweak use the old libgtk3 naming [11:24] so I guess it is because of some other dependency [11:24] I'll check on my vm [11:25] rodrigo_, http://paste.ubuntu.com/572136/ [11:25] rodrigo_, on a natty system [11:25] yes, I guess it's because of some other dependency [11:26] * rodrigo_ looks [11:27] rodrigo_, seb128: do you know whether bug 710901 is fixed in gnome 3? i. e. should we report this upstream, or investigate backporting the gnome 3 yelp? [11:27] Launchpad bug 710901 in yelp "search not implemented" [High,Triaged] https://launchpad.net/bugs/710901 [11:27] pitti, hmm, let me test it [11:28] pitti, we have the GNOME3 version in natty... [11:28] oh [11:28] pitti, it has just patched to build with gtk2 [11:28] pitti: later, as currently there is just one var RUN_SMOKETEST, that toggles everything. I would have to do a "even if RUN_SMOKETEST is false, still try one if you find epm" and for that I would have to check if there are other vaild reasons that disable the smoketest. For example the smoketest is disabled on i386/armel because it seems to currently fail there although the product is fine ... [11:29] pitti: seems to not work for me, but that doesn't mean anything, since I'm now with a half-updated system [11:29] Sweetshark: ok, so maybe better disable it by default, and provide an easy way to run it manually locally? [11:30] rodrigo_: I confirmed the bug; I overlooked that we already have the gnome 3 version [11:30] ok [11:30] rodrigo_: would you have some time to report this upstream and perhaps check if it's a missing feature or perhaps just a simple bug? [11:30] pitti: It is pretty easy to run manually, if you know how to find it ;) [11:31] Sweetshark: you know it, which is sufficient from my POV :) === cypher is now known as czajkowski [11:33] * pitti -> supermarket, bbl [11:33] * Sweetshark wonders however, how 3.3.0 for natty did build without epm ... === MacSlow is now known as MacSlow|lunch [12:14] is lp too busy? it's not building anything new in the gnome3 ppa [12:14] pitti, yes, doing it now [12:16] pitti chrisccoulson, re: Natty langpacks, do you think we could look at them next week to see if po2xpi builds the Firefox translations correctly? It would be great if we could ship FF translations in A3 [12:16] I'll be more than happy to help [12:16] oh, we're not shipping translations yet? [12:16] i didn't realise :( [12:16] i thought that was solved already ;) [12:17] chrisccoulson, you should read your inbox :P [12:17] heh [12:17] :) [12:18] dpm - are you referring to bug 715733 ? [12:18] Launchpad bug 715733 in po2xpi "XPI files are not correctly built for whitelisted languages" [Undecided,New] https://launchpad.net/bugs/715733 [12:21] chrisccoulson, yeah, that was one thing we encountered while creating the Lucid langpacks. It happened to be an LP bug, which is now fixed, so I've jsut marked that one as Invalid. For Natty, I think it's just a matter of testing if running the latest po2xpi version which included danilo's changes to adapt to the new upstream xpi layout produces valid FF translations [12:23] this can be done by downloading the FF translations from LP and running po2xpi locally in the same way it is invoked by langpack-o-matic, and see if the produced language tarballs look ok [12:24] pitti, hmm, git seems to have it implemented [12:26] pitti, can't install right now the latest yelp because of webkit broken package, so you see the same message in the terminal than in the bug? [12:38] dpm: yeah, I was planning to [12:40] pitti: Hello, were you notified about this build failure: https://launchpad.net/ubuntu/+source/gdm/2.32.0-0ubuntu9/+buildjob/2282264 ? [12:42] GunnarHj: hi [12:42] GunnarHj: I wasn't actually, because your name is in the changelog; looking [12:42] GunnarHj: seems transient, I'll retry the build [12:42] pitti: Ok. [12:43] GunnarHj: oh, it was already apparently [12:43] ah, no, ubuntu9, not ubuntu8, sorry [12:44] pitti, ok, the bug reporter is running 2.91.10, 2.91.90 does include the implementation, so should I submit a branch for upgrading it to 2.91.90? [12:44] GunnarHj: ah, that's powerpc; i386/amd64/armel built fine [12:45] rodrigo_: that'd be nice; we don't want to ship with a prerelease anyway, but upgrade to 3.0 final eventually [12:45] rodrigo_: thanks! [12:45] pitti, ok, submitting in a while [12:45] rodrigo_: I assigned the bug to you [12:45] ok [12:45] pitti: Right - for a while I was wondering where you read. ;-) [12:46] GunnarHj: I have a script to retry a package build on all arches, and it apparently picked the wrong version [12:46] pitti: check-mir chokes on the LO deps :( [12:46] \o/ empathy is back [12:46] Sweetshark: what else is missing? [12:46] Sweetshark: or do you mean it crashes? [12:47] it crashes [12:47] failing to parse the deps [12:47] pitti, oh, robert_ancell already added 2.91.90 to lp:~ubuntu-desktop/yelp/ubuntu, building it now and if it builds, I guess you can submit that [12:47] pitti: Special script, ok, I see. [12:48] rodrigo_: cool [12:48] Sweetshark: ah, output? [12:49] pitti, well, can't build because of broken libwebkitgtk-1.0-0 :( [12:49] pitti: https://pastebin.canonical.com/43948/ [12:49] is libwebkitgtk-1.0-0 broken for anyone else, or is it just me? [12:50] Sweetshark: uh, that's in python-apt, not even in check-mir itself [12:50] rodrigo_: how so? [12:51] pitti, libwebkitgtk-dev : Depends: libwebkitgtk-1.0-0 (= 1.3.12-0ubuntu1) but it is not going to be installed [12:51] Sweetshark: the Depends: line looks fine to me, hmm [12:51] rodrigo_: hm, just installed libwebkitgtk-dev on amd64, works fine here [12:52] rodrigo_: rodrigo_ do you have libwebkitgtk-1.0-0 installed? which versino? [12:52] pitti, no, not installed [12:52] rodrigo_, what does apt-cache policy libwebkitgtk-1.0-0 says [12:52] rodrigo_: what's "apt-cache policy libwebkitgtk-1.0-0" for you? [12:53] * pitti ^5s seb128 and takes a penalty card for snapping [12:53] libwebkitgtk-1.0-0: [12:53] Installed: (none) [12:53] Candidate: 1.3.12-0ubuntu1 [12:53] Version table: [12:53] 1.3.12-0ubuntu1 0 [12:53] 500 http://archive.ubuntu.com/ubuntu/ natty/main amd64 Packages [12:53] pitti, ;-) [12:53] hmm, I've got no libwebkitgtk packages installed [12:53] rodrigo_: what does this say: sudo apt-get install libwebkitgtk-1.0-0 libwebkitgtk-dev [12:53] it should point out the error more clearly [12:53] pitti: it seems not to like the "[!armel]" stuff [12:54] Sweetshark: in your current control, is lucene and epm gone now? [12:54] Sweetshark: ah, that's be an mvo bug then [12:54] pitti, complains about -common, but 'sudo apt-get install libwebkitgtk-1.0-0 libwebkitgtk-dev libwebkitgtk-1.0-common' works [12:54] werid [12:54] it uninstalls all the libubuntuone packages [12:54] and my typing sucks today [12:54] that's why it was complaining [12:54] * rodrigo_ installs [12:54] The following packages will be REMOVED: [12:54] banshee-extension-ubuntuonemusicstore gir1.2-ubuntuone-1.0 [12:54] libubuntuone-1.0-1 libubuntuone-dev libubuntuone1.0-cil [12:54] libubuntuone1.0-cil-dev libwebkit-1.0-2 libwebkit-1.0-common libwebkit-dev [12:55] pitti, ok, building the package now [12:55] pitti, seb128: you don't have the u1 packages installed I guess? [12:55] rodrigo_: ah, right [12:56] so, libwebkitgtk is the new name for the packages, right? [12:56] I think they got removed yesterday during dist-upgrade, and I accepted (deliberately) [12:56] :) [12:56] rodrigo_: seems so, yes [12:57] ok, the u1 packages need an upgrade then [12:58] go pitti! \o/ [12:59] pitti: with a :%s/\[\!armel\]//g on control in vim it only finds this: [12:59] pitti: https://pastebin.canonical.com/43952 [12:59] seb128: ? [12:59] pitti, (I was going to ask if we should try to get the new policykit but I noticed you synced it) [12:59] Sweetshark: cool, that looks fine; go! [12:59] seb128: :) [13:00] pitti: go what? dput it to main and be cruelly denied? [13:01] * Sweetshark makes sure to update the copy on chinstrap [13:01] Sweetshark: I mean "let's get this uploaded" :) [13:01] right, chinstrap -> pitti -> dput route again [13:01] * pitti thinks this is all just Sweetshark's s3kr1t plan to get 10.000 upload karma in one day [13:02] *hrhr* [13:02] * Sweetshark puts finger to the edge of his mouth: ONE MILLION!11! [13:05] uuh /me bows in awe [13:05] Sweetshark: if it's proportional to package size.. :) [13:07] well, since I dput via you arent you stealing all that karma anyway? [13:09] pitti: you should find the files on chinstrap [13:11] Sweetshark: no, I shouldn't [13:11] Sweetshark: I already get tons of false karma because it puts the langpack uploads on my account [13:12] LibO, Take III [13:13] * Sweetshark prays for it to work this time. [13:13] ... and tags ubuntu3. [13:15] I wonder what _rene_ will say later about me spamming the debian repo with ubuntu releases. [13:19] Sweetshark: don't you only apply the tags to the ubuntu branch? [13:19] chrisccoulson: can you please help my brain with an update of bug 663294? does this really affect our i386 firefox package, or only if you enable some build options? [13:19] Launchpad bug 663294 in gcc-4.5 "Firefox built with gcc-4.5 is a non-starter on i386 with -pie" [High,Triaged] https://launchpad.net/bugs/663294 [13:19] seb128, tedg is gone today? when is he back? [13:20] mterry, tuesday [13:20] mterry, he's at scale or something [13:20] seb128, hrm, the datetime prefs didn't get merged because I forgot to bzr add an image file. I guess we'll do a FF for it [13:20] mterry, hey btw [13:20] FFe that is [13:20] pitti - it affects the i386 package if we build with -pie (which we did in maverick, and the security team want) [13:20] mterry, talk to dbarth when he's back from lunch, he was talking to klatimmer about that [13:20] i've disabled it for now though [13:21] seb128, k [13:21] mterry, I would recommend you do a trunk snapshot with your work distro patched today if you can [13:21] Sweetshark: do you know what will happen with oo.o-hyphenation & co? will these be renamed to libo as well? [13:21] pitti: yes, do. but they will show up on the summary page (and on the SCM mailing list) anyway. [13:21] mterry, btw I assigned you a new crasher from yesterday's ted's tarballs [13:21] mterry, not sure if you saw [13:22] Sweetshark: wohoo, building [13:22] mterry, would be nice to try to handle that today as well if you can, with some luck it's easy to fix [13:22] seb128, would still require an FFe, though eh? And it's a 6k patch, with data files [13:22] 6k line patch that is [13:22] seb128, I saw, I can look at it [13:22] chrisccoulson: doesn't seem realistic for a3, does it? is there any progress on that actually (upstream?) [13:22] mterry, right, but I think the ffe should be easy to get [13:23] mterry, it's just a matter of asking and getting it approved [13:23] mterry, but we can't wait on ted to be back if we want it in a3 [13:23] pitti: I have not checked those yet. Those are likely in the *-10n* package right? [13:23] pitti - no, that's definitely not going to happen for a3. i will look at it again next week though [13:23] Sweetshark: no, these are a separate source [13:23] seb128, ah... hrm. OK. I'll work on that today [13:23] stupid 'bzr add' mistakes.... [13:25] mterry: hi, so you have the missing file? [13:25] dbarth, yeah, updated branch [13:25] mterry: what's blocking otherwise, a code review? do you want me to get that arranged for you to be able to land the code? [13:25] hi [13:26] dbarth, I don't know how much tedg was able to review before giving up on the image. It's a big patch (6k lines) [13:26] yeah, i saw [13:26] i don't think he fully reviewed it [13:26] he probably wanted to get it running first [13:26] mterry, seems ted worked later yesterday seeing when he did this review [13:26] Yeah. === tkamppeter_ is now known as tkamppeter [13:26] mterry, there was a stack of libdbusmenu and other indicator changes to land [13:27] kamstrup: ping? do you have time for a code review for mterry here? [13:27] dbarth, so a review would be good [13:27] mterry, so I wouldn't be surprised if he didn't review much yours [13:27] seb128, did they not land? [13:27] mterry, they did but it seems ted got his day full with those [13:27] i see [13:27] mterry, so it's likely yours came as a end of day thing [13:27] sure [13:28] mterry, can you start on the crasher while they start the review etc? [13:28] yeah, he reviewed it pretty late at night and he was probably packaging up for scale [13:28] seb128, yup [13:28] mterry, ideally we would have an upload for monday for datetime, today better if you can [13:29] dbarth: Feel free to comment about my stupid ideas regarding the lo-menubar btw ;-) [13:30] Sweetshark, what idea did you have? [13:30] Sweetshark: eh, just reading now [13:30] pitti - is there any way to stop a user from reporting a bug with apport-bug if we detect certain conditions exist on the system? === MacSlow|lunch is now known as MacSlow [13:31] ie, i want my apport hook to stop and tell the user that their system is broken [13:31] chrisccoulson: yes, you can check for the conditions in the package hook, and then set report['UnreportableReason'] [13:31] chrisccoulson: grep for Reason in /usr/share/apport/package-hooks, plenty of examples [13:31] pitti - excellent, thanks! [13:31] chrisccoulson: or have a look at the -- shock!! -- documentation: /usr/share/doc/apport/package-hooks.txt.gz [13:31] heh :) [13:32] i thought it would be quicker to ask you first ;) [13:32] oh, that's fine [13:32] i'm going to reduce the firefox bug count by quite a bit :) [13:32] * pitti just is proud that he actually took the time to write some docs about this stuff [13:33] chrisccoulson: what are you going to check for? grepping for "pr0n" in the history? :-) [13:33] pitti - not quite ;) [13:33] i want to check if /usr/bin/firefox has been diverted with dpkg-divert [13:33] seb128: LO and Nattys release schedules working together -- and make lo-menubar fit somewhere inbetween there .. [13:34] installing packages from ubuntuzilla does that, and it's our number 1 reported bug [13:34] chrisccoulson: sounds great [13:34] i've just added some cleanup code to our postinst script to fix peoples installs [13:34] Sweetshark, hum, aren't those scheduled fixed for a while? ;-) [13:34] schedules [13:34] Sweetshark, the lo-menubar was going to land this week it seems [13:34] kenvandine was on it yesterday [13:39] Sweetshark: but so 3.4 is not for natty, right? [13:40] i've asked aruiz to join [13:41] dbarth: no, 3.4 is not for natty as it final is scheduled after the natty release. [13:42] i'd rather try to get a module running on the version shipped with natty [13:42] Sweetshark, dbarth, hey hey [13:42] as i understand getting updates is what the interface doesn't provide [13:42] aruiz: hi [13:42] dbarth, yeah, that's the main problem, but there are others issues that it might not be possible to solve [13:43] dbarth: yes, there are some menus that are being populated the moment they popup, so their content is not static [13:43] polling for changes on user events may be an acceptable solution [13:43] ie, it's not "dumb" polling [13:43] dbarth, that's what ted suggested [13:43] seb128, see, I've already fixed the crash [13:43] ah, see ;) [13:43] we can use the on-about-to-show event on the appmenu indicator [13:44] seb128, and there's still 3 more hours! [13:44] dbarth, the problem is that on slow machines it wouldn't work properly [13:44] ah, ok, in this sense [13:44] mterry, excellent [13:44] mterry, do you need sponsoring for it? [13:44] seb128, only if it's not in desktop suite [13:44] * mterry checks [13:45] dbarth, other than that, there are some minor issues, things won't be perfect but they'll kinda work [13:46] seb128, I'm good [13:46] I'm working on the shortcuts at the moment [13:46] ok [13:46] mterry, great! ;-) [13:46] mterry, don't wait on ted's review since he's not there, just upload to natty [13:46] so, yeah, i'd prefer to secure a solution that would work as a module on the stable release shipping in natty [13:47] dbarth, I do prefer that as well, from the perspective that I want to deliver something you're happy with within schedule, but keep in mind that there might be some things that will not be possible [13:47] i feel that the workarounds required for doing that should be adressed in parallel, or post release and target 3.4+ [13:48] dbarth, I'm leaning towards pushing as far as I can with what's available through UNO for natty [13:48] and making it as decent as possible [13:49] aruiz: agreed [13:49] dbarth, I can certainly do that _if_ I don't have to deal with the insides of openoffice in this contract, and if it's okay to ship it even if it's not perfect [13:50] seb128: sorry for that upload, I should have unsubscribed sponsors instead of just leaving the bug like that [13:50] dbarth, in the long run I agree with Bjoern's approach [13:51] Sweetshark, does this make sense to you? [13:52] dbarth aruiz: understood. My point about upstreaming the stuff to master (3.4+) is mostly so that we try to do that as soon as the suff is reasonably stable. Squeezing it in on the last day before (upstream) feature freeze might get a bit hectic. [13:52] Sweetshark, if I have some time before we try to integrate it, I can clean up the code a bit and document things further, that'll make your work easier if you take it from there [13:53] aruiz dbarth: So going with a module/extension approach for 3.3 which might not be able to satisfy every detail, and going upstream for 3.4 to get the rest fixed sounds good to me. [13:54] we're all on board [13:54] nice! [13:56] aruiz: I already had a peek at the repo, the LO stuff was easy to understand, so I would rather need help on the dbus stuff (but that is nothing a little bit of tutorial reading would not fix) [13:56] Sweetshark: ok, so we're all on the same page [13:56] dbarth aruiz: yep, great! [13:56] Sweetshark, the dbus stuff is pretty straightforward, I will assist you with that [13:57] Sweetshark, installing d-feet might help you, it's an instrospection tool for dbus [13:57] aruiz: The minds of longtime LO/OOo hackers can be rather spoiled ;) [13:58] Sweetshark, tell me about it [13:58] X-) [13:58] hrhr [13:59] * Sweetshark will have a look at d-feet soon [13:59] seb128, OK, did you have other crashers queued up for me? [13:59] Sweetshark, :-) [13:59] mterry, in indicator-applications? no [13:59] seb128, but elsewhere, I remember you talking about not assigning stuff. [14:00] seb128, did that theme_changed_cb one get fixed? [14:00] mterry, there was some libdbusmenu crashers but quite some fixes landed this week so we need to see how that settle [14:00] oh no [14:00] mterry, https://bugs.launchpad.net/ubuntu/+source/indicator-application/+bug/708188 [14:00] Launchpad bug 708188 in indicator-application "softwares using libappindicator crash with SIGSEGV in theme_changed_cb()" [High,Triaged] [14:00] mterry, right, this one remains as something to fix for natty [14:01] mterry, you might want to sort indicator-datetime today and switch to crash fixes next week rather [14:01] seb128, sure [14:01] mterry, well, up to you, but I think those can wait a bit, it will also give us the w.e to see what crashes remaining after this week updates [14:02] dbarth, did you find a reviewer for my branch? [14:02] let's get the feature freeze rush to settle and see where we stand [14:04] kenvandine, howdy [14:06] pfff, my usb headset is not detected anymore [14:07] oops, wrong channel [14:09] w00t, just closed out lots of firefox bugs \o/ [14:12] hello everyone. QUestion: I made 2 releases yesterday using lp-project-upload, but none is appearing in Launchpad.Any idea how to debug what's wrong? [14:13] pitti: maybe you have a hint about ^? [14:13] (hi btw!) [14:15] hey nessita [14:16] nessita: the script didn't error out, I figure? [14:16] not in stdout/stderr at least. Is there any log file I can look at? [14:17] nessita, hey [14:18] hi seb128, how are you? [14:18] nessita, I'm fine thanks, how are you? [14:18] pretty good! happy that is Friday ;-) [14:18] ;-) [14:19] rodrigo_, does the nautilus patch need a leak fix? cf cosimo comment [14:20] pitti, bah the i386 retracer crashed again, I cleaned in .launchpadlib and restarted, let's see [14:20] hmpf [14:20] ok, works again [14:20] pitti: I ran it twice for the control panel, and the second time it didn't ask if I wanted to create the milestone. So that indicates that the milestone was created the first time, but is not showing in LP either [14:20] seb128: we might just put the rm -r into the cronjob.. [14:20] that's a weird issue [14:20] pitti, right, if it keeps doing that we should [14:20] seb128: I wonder if something in the cache file suddenly became arch specific [14:21] well it's a small issue but it's annoying [14:21] pitti, could be [14:21] bah [14:21] nessita: that smells like a LP bug then [14:21] 02/25/11 14:20:54: retracing #723712 failed with status: 99 [14:21] 02/25/11 14:20:54: transient error reported; halting [14:22] seb128: gateway timeout? I got those a lot yesterday [14:22] didrocks, is the launcher meant to appear when i hover over the BFB now? [14:22] pitti: any ideas how to debug? shall I do all the release 'by hand'? [14:22] or am i meant to click on it? [14:22] chrisccoulson: no, it's meant to appear, but there is one case it doesn't :) [14:22] nessita: if it doesn't throw any exception, I have no idea to debug :( [14:22] (it should fade to be honest) [14:22] I just fixed it in trunk [14:22] didrocks, oh. it never seems to work here ;) [14:22] chrisccoulson: hum? really? [14:22] like, press super [14:22] nessita: you could run the launchpadlib commands in lp-shell and see where it fails [14:23] release [14:23] not, hover the bfb [14:23] nessita: and then report a bug against launchpadlib with that (and the command you tried to do) [14:23] now* [14:23] didrocks, i didn't try super yet [14:23] chrisccoulson, you need to hit the corner [14:23] pitti: right [14:23] oh, it's appearing now [14:23] chrisccoulson, it doesn't stick otherwise [14:23] chrisccoulson: "super" is just a way to reset it :) [14:23] seb128 - i was hitting the corner. i wasn't even seeing it fade in before ;) [14:23] didrocks, workaround for the win ;-) [14:23] right :-) [14:24] will be fixed with Monday's release [14:24] still working on some false positives :) [14:24] cool [14:29] dbarth, sorry, dropped of the net, so I'll risk repeating myself: is there a reviewer lined up for datetime? I think klattimer has a branch too. Would you prefer doing a tarball release today or just distropatch everything? [14:31] mterry_, you can add bug #723463 to your list of things to check if you can [14:31] Launchpad bug 723463 in indicator-appmenu "indicator-applet-appmenu crashed with SIGSEGV in dbusmenu_menuitem_send_about_to_show()" [Medium,New] https://launchpad.net/bugs/723463 [14:31] mterry_, he said to me that you should make a tarball earlier [14:31] mterry_: MacSlow is reviewing your branch now [14:31] seb128, cool [14:32] dbarth, k, ( MacSlow, thanks and sorry for size!) [14:32] dbarth, how about karls? [14:32] mterry_, np yw... everybody has a 6k+ patch to go in... but only per year ;) [14:32] ted didn't know the status of that last night [14:33] MacSlow, guh, it's only Feb [14:33] hey kenvandine [14:33] mterry_, guess you get to relax for the rest of the year [14:33] :) [14:33] hey seb128 [14:33] mterry_, yeah... used up all you goodie-tokens right at the start ;) === warp11 is now known as warp10 [14:33] kenvandine, !!! [14:33] kenvandine, don't tell that to mterry_ ;-) [14:34] hehe [14:34] kenvandine, how are you? [14:34] aw, I already bought a recliner too [14:34] tired, very tired [14:34] kenvandine, do you want to do the notify-osd update or should I do it? [14:34] i'll do it [14:34] thank [14:34] kenvandine, great work yesterday [14:34] kenvandine, take it easy today, it has been a crazy week and you deserver to relax a bit [14:35] seb128, ted's 0.3.99 release of dbusmenu caused every click on every menu item in any of the indicators or appmenu crash unity panel and indicator applet [14:35] that was fun to debug last night [14:35] kenvandine, was ted still around when you debugged it? [14:35] of course he couldn't reproduce it for several hours [14:35] yeah [14:35] we fixed it, distro patched it in the upload [14:36] it never got uploaded like that [14:36] kenvandine, ok, nice catch ;-) [14:36] seb128, he never got a chance to debug the problem with menus for gnome-bt [14:37] but he will [14:37] kenvandine, the update seems to work fine, gnome-bt refresh issues seem to be fixed as well [14:37] it seems better [14:37] well, seems to work now [14:37] but not completely fixed [14:37] there was an indicator-application crasher, mterry just fixed it [14:37] if i click turn off bluetooth, the menus change [14:38] whoops, i meant they don't change [14:38] it turns off [14:38] but the menu still says "Turn off bluetooth" [14:38] instead of On [14:38] and the submenus are still missing [14:38] seb128, can you confirm that? [14:39] kenvandine, I will tell you once I get mterry's update [14:39] indicator-application is just crashing now [14:39] on a null variant [14:40] ok, it built [14:40] let me grab the deb on launchpad [14:41] seb128, yeah, i got on my desktop box after upgrading it this morning [14:41] but not my laptop [14:43] kenvandine, no, refresh works fine [14:43] can't confirm your bug [14:43] when I turn bt off I get 3 lines only in the indicator [14:43] oh, got the issue now and the indicator crashed [14:44] #0 0x0060a099 in dbusmenu_menuitem_property_set_variant () from /usr/lib/libdbusmenu-glib.so.3 [14:44] No symbol table info available. [14:44] #1 0x0060a36d in dbusmenu_menuitem_property_set_int () from /usr/lib/libdbusmenu-glib.so.3 [14:44] great, new libdbusmenu crash [14:44] it's going to be a fun week tracking those [14:45] kenvandine, so it works a few times then get confused and crahs [14:46] yeah, some signal getting missed somewhere [14:46] i don't get a crash though [14:47] and i couldn't find similar behavior anywhere except gnome-bt [14:48] kenvandine, can you confirm that if you do on off a few times it crashes? [14:48] no, no crashes [14:48] i've done it dozens of times [14:48] pitti, bah the retracers crash every second retrace [14:48] pitti: is there any convention for naming backports? you proposed 3.3.1-1ubuntu3~maverick while I see older releases following this scheme: 3.3.1-1maverick3. [14:49] seb128: on that weird lp connection bug? [14:49] Sweetshark: append ~maverick or ~maverick1 (which one is a matter of taste) [14:49] pitti, on lp errors, not sure if it's the weird connection, I don't have enough buffer to scroll up after the launchpad noise [14:50] Sweetshark: appending just "maverick" instead of "~maverick" is actively wrong, though [14:50] seb128, yes, it needs it, talking to cosimoc about it, will push to git and submit a new patch for the package [14:50] rodrigo_, ok, I'm not uploading for now then [14:50] seb128, ok, just a second [14:51] rodrigo_: how did the yelp build go now? [14:51] rodrigo_, no hurry, it's just an issue when the svg loader is broken, not in normal use [14:52] seb128: FYI, I commented out MAILTO in retracers; those dchroot messages get on my nerves, let's wait until the RT gets fixed [14:52] pitti, it builds, but seems to not find the docs, so investigating [14:52] and amongst those I won't see actual error messages anyway [14:52] pitti, ok, I didn't notice I set up a filter for those the day they started ;-) [14:53] pitti: teh releases do not have a plain maverick appended, they have ubuntu _replaced_ by maverick [14:53] oh, come on [14:53] pitti, stupid retracers crash every 1 retrace now on launchpad error [14:53] *sigh* [14:53] I keep restarting them manually for now to clean the queue [14:54] but we will need to investigate next week I guess [14:54] seb128: transient? or permanent? [14:54] transient [14:54] it retraces one or two then crash [14:54] I've been getting tons of timeouts in the web ui as well today [14:54] restarting works on one or two and it crashes [14:58] MacSlow, btw, my branch does the UI for the prefs, but some of the preferences themselves don't change anything in the indicator. That's hopefully largely fixed by klattimer's branch [14:59] pitti, well at least the stacktrace are quite useful since yesterday [14:59] pitti, the new gdb works great it seems [14:59] (only the indicator preferences that deal with showing/hiding parts of the indicator don't work) [14:59] oh, great! [15:00] seb128, indicator-applet-appmenu doesn't even render at all in my gnome-panel... (looking at that bug). I assume that's something unique to me [15:00] mterry, so how do I best test this then? [15:00] mterry, I'm under unity so not sure [15:00] MacSlow, well, the first panel works fine. The left half of the second panel works fine. The right half you'll have to confirm that I'm twiddling gsettings by using the gsettings command [15:00] I don't want to start a guest session now, intel tends to crash on user switching [15:01] ok, I'm out for ~2 hours [15:01] I will read backlog and catchup for a bit when I'm back [15:02] mterry, I'll get back to you when I've sorted out the build-dependencies for your branch... atm fighting broken packages on my system [15:02] MacSlow, bummer :( [15:02] mvo: the aptdaemon you re-uploaded to maverick-proposed: all natty tasks are still open for that; is it fixed there? [15:03] bug 665572 even says "incomplete" [15:03] Launchpad bug 665572 in aptdaemon "Unhandlable programming error (LockFailedError) - lock.status_lock doesn't contain a valid path" [High,Incomplete] https://launchpad.net/bugs/665572 [15:04] seb128, https://code.launchpad.net/~rodrigo-moya/ubuntu/natty/nautilus/fix-leak/+merge/51307 [15:04] pitti: should be, let me double check [15:09] mterry, ok... building works... any chance to install it outside /usr (setting $PREFIX leaves "Indicator Dir" untouched pointing to /usr/lib/indicators/5) [15:10] pitti, ok, can't find the main docs, but have tested 2.91.90 with ghelp:gnumeric and searching works, so I think it's ok to go [15:10] cool, thanks [15:10] pitti, does gnome-open ghelp:gnome-help works for you? [15:10] rodrigo_: did you change anything in the branch? (it's yelp/ubuntu)? [15:11] pitti, no, nothing [15:11] rodrigo_: "The URI ‘ghelp:gnome-help’ does not point to a valid page." [15:11] yeah, seems we are missing that [15:11] rodrigo_: ok, will sponsor that then; thanks for the investigation! [15:11] in what package is the user doc? [15:11] MacSlow, it's probably getting that from libindicate's .pc file or some such. I haven't tested with it outside of /usr [15:11] gnome-user-guide is installed here, but the docs are not found [15:12] pitti, np [15:14] oh, it's ghelp:user-guide [15:14] mterry, ok... "sudo make install"ed it [15:15] mterry, so how do I make it reaload with the new one? [15:15] rodrigo_: that works [15:15] no idea then why yelp loads ghelp:gnome-help [15:15] have we changed anything? [15:16] MacSlow, there are two things to do. One is to restart the unity-panel-service. The other is to restart the indicator-datetime-service. [15:16] MacSlow, then you run indicator-datetime-preferences to get my sweet prefs dialog (or open it from the indicator) [15:21] mterry, I can't start /usr/libexec/indicator-datetime-service... [15:22] mterry, https://pastebin.canonical.com/43969 [15:23] I know he's not here, but robert_ancel, omg, F1 opens help in the blink of an eye now! [15:23] rickspencer3, wow... how in the hell did he do that! [15:23] i've always wondered why it was so slow [15:23] MacSlow, oh, huh. should be /usr/lib/indicator-datetime/indicator-datetime-service. You may be best served by uninstalling and then building a deb package out of it [15:24] MacSlow, not sure if that's your problem though [15:24] there's also no binary "indicator-datetime-preferences" in the build-directory [15:24] kenvandine, it was slow because it used to use gecko ;) [15:24] MacSlow, hrm... [15:24] although, i've got no idea why it was *that* slow [15:25] chrisccoulson, that was the only reason? geez [15:25] MacSlow, copy the debian directory from natty's package and use that at least. That way we can rule out build or installation issues [15:25] mterry, btw... I don't have a fully updated system, because I'm stuck with nvidia-current, which forces me to hold back a lot of updates [15:25] kenvandine, yeah, i'm still not sure why it was that slow. it was even slower than firefox, which doesn't make sense ;) [15:25] MacSlow, ok. well, grab the debian directory from whatever version of indicator-datetime you do have === jcastro_ is now known as jcastro [15:27] mterry, you mean using the debian directory with your branch to build a proper .deb?! [15:28] MacSlow, yeah [15:28] MacSlow, that's how I test anyway [15:28] MacSlow, I'm less familiar with the manual method [15:32] mterry, I can't build a package with this... I'm getting errors from dpkg-source I've no clue about... [15:32] MacSlow, ok. well let's continue with the manual way then. Let me re-examine your pastebin [15:33] MacSlow, I believe I've gotten that before erroneously when testing with restarting services and such. There's a trick to make it not timeout [15:34] mterry, do you know the trick? [15:34] MacSlow, working on it. I have to look it up [15:34] ok [15:34] kenvandine, do you remember the indicator debugging bash profile bits? [15:35] mterry, one sec [15:35] . /usr/share/libindicator/80indicator-debugging [15:36] MacSlow, ^ [15:36] kenvandine, thanks [15:36] no problem [15:36] MacSlow, you may want that in your ~/.profile [15:36] kenvandine, mterry: looking... [15:36] i don't know how well they work :) [15:36] i have mixed experiences [15:36] kenvandine, the not-crashs-when-no-one's-looking one works in my experience [15:36] kenvandine, I already did a "sudo make install" so not much more harm can happen now :) [15:37] MacSlow, with unreviewed code no less! [15:37] MacSlow, ugh... i never do that with anything :) [15:37] * mterry adds MacSlow's machine to his botnet [15:37] * kenvandine always tests from packages! [15:37] * MacSlow feels like an idiot [15:45] mterry, kenvandine: still getting the same error when trying to start indicator-datetime-service [15:46] MacSlow, you're running the new indicator (i.e. you restarted unity-panel-service? [15:46] yes [15:46] mterry, I see a clock in the panel [15:47] k [15:47] mterry, and I don't have the normal "indicator-datetime" package from natty installed [15:48] MacSlow, try: killall indicator-datetime-service; INDICATOR_ALLOW_NO_WATCHERS=1 /usr/lib/indicator-datetime/indicator-datetime-service [15:48] mterry, there's not binary called /usr/lib/indicator-datetime/indicator-datetime-service [15:49] MacSlow, oh ssorry, forgot your prefix was different [15:49] killall indicator-datetime-service; INDICATOR_ALLOW_NO_WATCHERS=1 /usr/libexec/indicator-datetime-service [15:49] MacSlow, ^ [15:51] https://pastebin.canonical.com/43972/ [15:52] mterry, but trying to start /usr/bin/indicator-datetime-preferences gives me... [15:52] https://pastebin.canonical.com/43973 [15:52] chrisccoulson: I guess there aren't news from mozilla about global menu? [15:52] MacSlow, ah... you're not using the latest gnome-settings-daemon [15:53] MacSlow, but at least the service is running [15:53] pitti - not yet, and the guy i've been talking to is not online atm :( [15:53] mterry, I guess that's one of the things I can't update to [15:53] MacSlow, is it possible for you to usse the latest g-s-d in natty? [15:53] in any case, the menu is broken because of dbusmenu, so i wouldn't upload it today ;) [15:53] mterry, because I've to stick with nvidia-current [15:53] but this is so goddamn depressing, having my work just sitting there :( [15:54] chrisccoulson: ok, thanks [15:54] MacSlow, the chain of dependencies for that must be odd then [15:54] i almost wonder if i should JFDI ;) [15:54] mterry, I'm trying to see if I can just grab that one... but I'm also having very odd mouse-click issues here right now [15:55] MacSlow, :-/ not your day [15:56] mterry, oh... I have these sort of issue all the time... not just today [15:56] :( [15:56] mterry, dev on a in-dev platform is asking for it [15:57] MacSlow, huh? [15:59] mterry, ok... was able to pull g-s-d [15:59] I'll restart that now [15:59] that didn't go well [15:59] is there a way to tell compiz that i don't want windows overlapping workspaces to appear on the next workspace? it's a weird experience coming from metacity and it breaks the unity panel auto-hide [16:01] chrisccoulson: release meeting starting now; I'll ask about swt-gtk [16:01] pitti - excellent, thanks [16:01] fta: No, that is an integral part of the design, it's how viewports work [16:01] hmmm, what happened to evolution? it's gone crash crazy this afternoon [16:01] mterry, success... at last [16:02] Amaranth, d'oh! :( [16:02] MacSlow, yay! [16:02] fta: We used to have support for virtual desktops (and the code is even still there) but afaik it last worked in 7.04 and you can't get expo, wall, cube, etc with it [16:03] chrisccoulson: would you mind opening a bug against euca to drop swt-gtk? then we can track this better [16:03] MacSlow, I don't know when you grabbed my branch, but I had made one last commit this morning removing the "week starts on" preference (as it won't work at all with current gtk calendar -- they ignore that preference in favor of locale) [16:03] chrisccoulson: (sorry, busy in meeting now); please add a rationale about supportability and xulrunner, etc. [16:03] Hi everyone. I am looking for anti-virus software for Ubuntu 10.10 Desktop? any suggestion? thanks [16:03] pitti - yeah, can do. (or move it to universe) ;) [16:03] chrisccoulson: or that :) [16:04] mterry, "Show a clock in the panel" has no effect... it's always shown [16:06] MacSlow, yup [16:06] MacSlow, that's one of the 'show preferences' that doesn't have code on the indicator side [16:07] Amaranth, hm, i don't want virtual desktops, do I? i often push my windows slightly off screen on the right, or bottom right when i need more visible space on a workspace (been doing that since fvwm 2 decades ago). but now, the off-screen parts appear on the next workspace, beside being ugly, it makes the problem worse, as it's even visible space there, but it also prevents the panel from raising [16:07] MacSlow, some of those are fixed by klattimer's branch (but not the 'show clock' and live updating of locations) [16:07] mterry, apart from that everything renders fine and acts correctly on the chagnes I make in ther dialog right away [16:07] Amaranth, +less [16:14] MacSlow, afk for like 20m [16:14] mterry, still looking through the code... i'll write a summary in the comment on the merge-proposal [16:29] MacSlow, back [16:38] MacSlow, ah, good leak finds, thanks [16:41] pitti: ping? [16:41] Sweetshark: o/ [16:42] MacSlow, fixed. who does the actual merge? [16:44] pitti: we currently separate libreoffice and libreoffice-l10n, but as both are generated from the same source I shouldnt need another branch for l10n, right? [16:44] (apart from control file regeneration) [16:44] Sweetshark: no; AFAIK it's just changing the name in debian/changelog, regenerating control file, debuild -S [16:45] Sweetshark: their orig.tar etc. shoudl be identical [16:46] on a down note: the arm build failed because gcc is too stupid (or too clever) for itself ... [16:47] ? [16:47] I've seen the error, but it looked like an ICE [16:48] pitti: yes [16:53] mterry: did you get a review, are things all good for sealing a release of i-datetime? [16:54] dbarth, MacSlow approved. but now I'm looking at getting klattimer's datetime stuff in too [16:55] mterry: he pushed rev 67 he told me [16:56] dbarth, I guess I don't know what that means [16:56] dbarth, who merges my branch now that it's approved? [16:57] dbarth, klattimer's stuff needs some more baking and moving-parts-settling it seems. I think mine was the more FF-important bit [16:57] that's the issue, as foundations is a bit empty today [16:57] mterry: can you roll a release from trunk now? [16:58] sorry, merge in your changes [16:58] i'll ask karl to do the same for his [16:58] dbarth, I'm not in ~indicator-applet-developers [16:58] ah [16:58] good night everyone, have a nice weekend! [16:59] pitti, by! [17:00] enjoy pitti [17:01] pitti: have fun! [17:07] dbarth, talking to klattimer, he's fixing up his branch to now apply on top of the new trunk. his branch also needs a new libido. is that lined up for release too? [17:08] mterry: ok, great, njpatel is your man now, as i'll take the road in a minute now [17:08] dbarth, ok [17:08] * mterry hugs njpatel [17:09] * njpatel hugs mterry [17:09] Why are we doing this? [17:09] MacSlow, btw, thanks for your review! I know that was a lot to dump on ya [17:09] njpatel, you just got volunteered to help shepherd the indicator-datetime stuff to release [17:09] njpatel, I just had a giant branch land, and klattimer is fixing up his merge for it too [17:09] njpatel, his merge requires a second merge on libido which will need a release too [17:10] njpatel, and both will need FFe I suppose. I can help with that bit though [17:14] njpatel_, you there? [17:15] mterry, hey [17:16] sorry, connman decided to crash === njpatel_ is now known as njpatel [17:16] njpatel, how much did you get? [17:16] the last sentence only :/ [17:16] can you pastebin what you said please? [17:17] mterry, ^ [17:18] njpatel, http://pastebin.ubuntu.com/572261/ [17:19] njpatel, FFe bug filed but not yet official: https://bugs.launchpad.net/ubuntu/+source/ido/+bug/725121 [17:19] Launchpad bug 725121 in indicator-datetime "Feature Freeze Exception for indicator-datetime" [Undecided,New] [17:19] mterry, ah, thanks [17:19] hold up [17:20] (i didn't subscribe the team yet, just doing paperwork ahead of time) [17:36] seb128, ok, talked to njpatel. I'm going to fix up klattimer's branch to apply to new trunk, njpatel will review and possibly publish tomorrow. FFe (started in above link) will hopefully get granted on monday [17:37] mterry: there is still no more delegation discussed, but yeah, just emphasize on why the FFe is important [17:37] then, it should be good [17:38] didrocks, delegation? [17:40] mterry: yeah, the release team is giving FFe [17:40] mterry: but in the past, for some part of the archive, some non release team member can give FFe [17:40] didrocks, ah [17:40] like seb128 in GNOME, me in UNE [17:40] and so on… :) [17:40] "the good ol' days" [17:40] right ;) [18:01] kenvandine, did you break evolution? ;) [18:01] chrisccoulson, not that i know of :) [18:01] chrisccoulson, what's up? [18:02] kenvandine, oh, it crashes every couple of minutes since i updated my machine earlier ;) [18:02] humm [18:02] haven't seen that [18:02] all i changed in evolution was the desktop file [18:02] but i did change evolution-indicator [18:03] kenvandine, http://paste.ubuntu.com/572279/ [18:04] i haven't got all the symbols installed just yet though [18:04] happens pretty much every time it scans for new mail ;) [18:04] didrocks, dbarth - do you have a nice summary of the actual packages with updates still left to land for A3 (ie FFEs)? - is the bug mterry the only one anticipated? [18:04] (every 2 minutes) [18:04] skaet_: there will be nux and unity of course, maybe libunity as well [18:05] skaet_: but it's not a FFe for alpha3 [18:05] just bug fixes [18:05] (maybe a small fix to the compiz decoration as well, so compiz to the list) [18:06] chrisccoulson, ok, that does look like my fault :) [18:06] heh :) [18:07] not sure why it is using libnotify1 though [18:07] so not really my fault :) [18:07] chrisccoulson, please file a bug, i'll look at it [18:07] urgh - Complete report 172.3MB [18:07] perhaps i won't be submitting it just yet ;) [18:07] that's surely a joke isn't it? [18:08] didrocks, thanks. bug fixes are welcome, regressions can please stay in local development trees. ;) [18:08] ugh [18:08] no idea [18:08] :) [18:08] skaet_: yeah, but there is never the joy of delivering a bug fix without a regression :-) (kidding) [18:08] chrisccoulson, i think i know why... sigh [18:09] chrisccoulson, just file the small version of the bug :) [18:09] :) [18:09] kenvandine, the full version is uploading fairly quickly, i think apport must have been lying ;) [18:12] kenvandine, oh, bug 725102 :) [18:12] Launchpad bug 725102 in evolution-indicator "evolution crashed with SIGSEGV in notify_notification_new()" [Medium,New] https://launchpad.net/bugs/725102 [18:13] chrisccoulson, thx [18:17] re [18:17] mterry, ok great [18:23] chrisccoulson, ok, uploading fix now [18:23] thx! [18:23] kenvandine, excellent, thanks :) [18:26] chrisccoulson, do you know about the loss of appmenu menus in firefox? [18:27] chrisccoulson, says it is incompatible with firefox 4.0b12 [18:27] kenvandine, have you updated globalmenu-extension from my PPA? [18:27] i wonder if i disabled your ppa [18:28] i guess that explains it :) [18:28] * kenvandine adds ppa aain [18:28] again [18:28] heh [18:28] that should fix it :) [18:29] of course, if we were shipping it by default, we wouldn't have this issue (as I would just build it from the firefox source tree) [18:29] * chrisccoulson sighs [18:30] yay, thx chrisccoulson [18:31] speak of that... seb128 mind reviewing lo-menubar again? [18:31] kenvandine, it was ok yesterday, do you need another review? [18:31] it's in sourceNEW [18:32] kenvandine, ok [18:32] thx [18:32] seb128, remember you are betting on me being first :) [18:33] well, i think that's a given now ;) [18:33] unless I just go ahead and upload it [18:33] which i'm getting close to doing ;) [18:34] chrisccoulson, they said today right? [18:34] kenvandine, yeah, but that's not going to happen now. i haven't heard anything from the guy i'm talking to and he's not online either [18:35] if ever we need a reason to switch to chromium, then this is it ;) [18:36] :/ [18:41] kenvandine, ok, source NEWed [18:41] thx [18:41] yw [18:55] ok, time for dinner her [18:55] here* [18:55] see you next week! [19:05] if someone wants to review small bugfixes to evolution and evolution-exchange: https://code.edge.launchpad.net/~mathieu-tl/evolution-exchange/lp694884/+merge/51219 and https://code.edge.launchpad.net/~mathieu-tl/evolution/2011-02-25/+merge/51352 [19:12] does anyone know what's going on with the amd64 builders? everything is failing now [19:14] oh, it's not just amd64 now [19:19] bryce_, X crashes on startup (in evdev_drv) [19:20] fta_, file a bug, we'll investigate when we get time [19:21] fta_, a full backtrace via gdb or apport is the most useful thing to gather in that case [19:21] bryce_, difficult, i'm on my tablet right now.. http://pastebin.ubuntu.com/572317/ [19:21] fta_, yeah hard to guess from that limited trace [19:22] i know, but nothing in /var/crash for it [19:22] fta_, presumably something with the utouch patches that went in earlier this week [19:22] pid of X ; gdb /usr/bin/Xorg ; # attach ; bt full [19:22] can't find the evdev-drv dbg deb.. [19:23] xserver-xorg-input-evdev-dbg [19:24] nope, E: Unable to locate package xserver-xorg-input-evdev-dbg [19:24] same for -dbgsym [19:25] huh, yeah you're right no -dbg for -evdev. weird [19:26] tried a downgrade, -6 gives me another crash :( [19:26] need help... i'm new to ubuntu [19:26] well the bug is probably in xserver code actually [19:26] is this the right channel? [19:27] ssk_the_gr8, sorry, no [19:27] ssk_the_gr8, you want #ubuntu [19:27] where should i go... [19:27] k thanx [19:31] bryce_, http://paste.ubuntu.com/572320/ [19:35] fta_, ok looks better [19:38] fta_, not sure offhand, I probably need to show this to cnd since I suspect it's that the input device logic has changed a bit so as to accomodate multitouch devices, and it seems to have adversely affected your particular device [19:39] oh speak of the devil [19:39] I was tipped off [19:39] you've been had! [19:40] heh [19:40] cnd, http://paste.ubuntu.com/572320/ [19:41] fta_, btw I'm assuming this is a system+device that's been running natty for some time, and not just a fresh upgrade or installation? [19:42] fta_, i.e. it's a recent regression? [19:42] bryce_, yep, removed my wireless mouse, it works [19:42] bryce_, yes, it's a recent regression [19:42] this box is using natty since day 1 [19:42] fta_: can you file a bug report: https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-input-evdev/+filebug [19:43] assign it to me, username "chasedouglas" [19:43] actually, might be better to use ubuntu-bug xserver-xorg-input-evdev [19:43] that'll automatically attach logs and such [19:43] bryce_, ahh, sure [19:43] now that i have my desktop back, i guess i can file the bug.. providing i find another mouse somewhere in the house :P [19:43] fta_, also attach your backtrace for him [19:43] fta_: can you also attach the gdb log file? [19:44] fta_: I will to look at this today [19:44] cool [19:44] thanks cnd [19:44] give me a minute [19:45] ahhh [19:45] better than an android virtual keyboard [19:57] cnd, bryce_: bug 725202 [19:57] Launchpad bug 725202 in xserver-xorg-input-evdev "X crashes on startup in evdev_drv" [Undecided,New] https://launchpad.net/bugs/725202 [19:58] fta: thanks! [20:30] cnd, as i said above, the previous version (-6) also crashes, for another reason [20:31] fta, oh!, I missed that [20:32] fta, did this bug occur before monday? [20:33] cnd, 1st time today. but i had X held for a while by the nvidia-current driver, which just appeared today [20:34] fta, does that mean you have seen crashes before with the previously held X version? [20:34] cnd, nope [20:34] ok [20:36] fta: can you try this: [20:36] switch to a VT [20:36] $ sudo stop gdm [20:36] $ sudo Xorg :0 [20:37] see if it crashes [20:37] when you have the device plugged in of course [20:37] $ sudo start gdm [20:37] to get back to X afterwards [20:38] cnd, tried that several times already, it always crashed [20:38] fta, ok, that's good for debugging at least [20:39] fta, please install valgrind [20:39] then in the VT, run: [20:39] i have it [20:39] $ sudo valgrind --tool=memcheck Xorg :0 > valgrind.log 2>&1 [20:39] i need to kill my session then.. [20:39] attach the file to the bug [20:39] ok [20:39] thanks for the hard work :) [20:52] cnd, got it, along with a /var/crash/_usr_lib_valgrind_memcheck-x86-linux.0.crash [21:00] chrisccoulson, didn't take long: Title: evolution crashed with SIGSEGV in notify_notification_new() [21:01] fta - a fix has been uploaded, but the builders are screwed [21:01] chrisccoulson, good, which package is that? i'll rebuild it locally, i'm stuck [21:01] fta - evolution-indicator [21:02] chrisccoulson, ok, thanks [21:04] chrisccoulson, 0.2.14-0ubuntu3 ?? it ftbfsed on i386 & amd64 [21:04] GNOME packages (mutter, gjs,...) get automatic grants for feature freeze? [21:04] fta - yeah, that's the one [21:04] bigon - yes, as long as they're already on unstable series i think [21:04] please don't upload anything right now though :) [21:05] alright [21:05] ah, ok [21:05] why? [21:05] fta - the failure is because the builders are broken [21:05] that's answer my question I guess [21:05] bigon - http://identi.ca/launchpadstatus [21:06] alright [21:07] cnd, did the valgring logs help? [21:07] fta, hmmm... it really hurts that the archive hasn't generated dbgsym packages for evdev like it's supposed to [21:07] so yes, the logs help [21:08] but I can't correlate the valgrind issue to source code right now [21:08] fta, let me build you a special package for testing [21:09] fta, actually, it does help some [21:09] it kind of looks like a simple off by one error [21:09] let me look back at the code again [21:18] chrisccoulson, confirmed, evo crash gone [21:28] fta, that was from my fix in evolution-indicator? [21:28] fta, ah, i read back... it is... great, thanks for confirming it fixed it [21:29] kenvandine, thanks for the fix :) [21:29] b'ah, couchjs is broken *again* [21:29] grrrrrrrrrrrrrrrrrrrrrrrrrr [21:30] fta, i still don't know why it wasn't crashing here... but it was an obvious problem [22:03] i think i'm in love with unity 2d [22:16] fta, I've had to fix pkg-create-dbgsym to be able to create test packages for you [22:16] sorry for the delay [22:20] cnd: what was wrong with it ? [22:20] lifeless, https://bugs.launchpad.net/ubuntu/+source/pkg-create-dbgsym/+bug/725274 [22:20] Launchpad bug 725274 in pkg-create-dbgsym "pkg-create-dbgsym doesn't recognize linux-any architecture" [Undecided,New] [22:20] ah phew [22:20] I've got a branch pushed, just doing the ffe stuff [22:20] i thought it might be related to the current incident with the builds farms [22:22] actually, is an ffe needed for a bug fix? [22:23] I don't think it is... [22:28] if I need to upload a package to main, do I still subscribe ubuntu-sponsors? [22:28] or do I subscribe some coredev specific team? [22:28] cnd, oh, that's why the deb is an arch all then. [22:28] it's weird [22:28] fta: ? [22:28] which deb is an all? [22:29] xserver-xorg-input-evdev-dev_2.6.0-1ubuntu6_all.deb [22:29] _all [22:29] cnd: are you asking WRT feature freeze? [22:29] fta, that's just the dev package [22:29] not the real one [22:29] oh [22:29] micahg, it's a bug fix to pkg-create-dbgsym [22:29] cnd: bug fixes are fine after feature freeze except when frozen for a milestone or past final freeze [22:30] micahg, thanks [22:30] n-m, wrong copy/paste [22:31] cnd: and in those cases, you can ask the release team if you think it's urgent [22:43] hm, found a regression in unity [22:44] mplayer in fullscreen mode no longer accepts keyboard actions [22:52] fta, it doesn't help that you're on i386 either :) [22:52] I had to go find a build resource to get you test packages [22:55] I'm getting there though... [23:00] fta, please install the two evdev packages at http://people.canonical.com/~cndougla/utouch/ [23:00] then do the valgrind dance [23:00] and attach the output to the bug again [23:00] thanks! [23:12] cnd, ok, will do [23:34] fta, thanks for the logs, it does look like my patches are causing stack corruption [23:36] cnd, ok, do you have enough info now? [23:36] fta, I think so [23:37] good [23:37] on my side, i had some fun controlling my desktop from my android tablet ;) [23:37] fta, I think I know what's wrong [23:37] I'll try to spin a new package for you [23:38] it's getting late here (paris) [23:39] fta, ahh, I'll follow up on the bug report [23:39] you can test later [23:39] thanks for your help today though! [23:39] sure, thanks