[05:06] good morning [05:08] Hey there didrocks. [05:08] hey RAOF, how are you? [05:08] I'm alright. [05:09] It's a pity the weekends are so short, though :) === _LibertyZero is now known as LibertyZero [05:09] heh, got 3 days for once. Was needed :) [05:10] Awesome! [05:23] oh no! the French are coming!! ;p [05:23] Morning didrocks. [05:23] hey TheMuso ;) [05:23] vish: heh, be prepared! [05:48] Good morning [05:49] Good morning pitti! [05:49] hey RAOF, how are you? [05:50] I'm ok :). [05:50] Hows about you? [05:50] hey pitti! had a nice week-end? [05:50] hey didrocks [05:50] we did, indeed! my niece got married, so we went to Dresden again [05:51] oh, excellent :) [05:51] how about you guys? [05:53] Cool! A fun wedding? [05:53] it was indeed! they picked nice places for the ceremony (a castle), and the party [05:54] was nice! 3 days week-end was needed. Just the time to relax and do some exercice + enjoying the feist in Lyon (fireworks) [05:54] http://piware.de/fotos/Hochzeit-Henriette-Andre-Juli2011/ :) [05:54] didrocks: recovered a bit from last week's stress then? [05:55] pitti: not really stressful (well a lot of questions/bashing on unity, ok…), but crazy hours, yeah. Totally recovered now! [06:03] pitti: ok, so no need for ping you on the apport dpkg issue from Raphael :-) [06:03] I did some email catchup on Friday [06:03] we discussed it at RMLL FYI ;) [06:10] pitti: did you see my g-p-m merge request? [06:11] jbicha: yes, thanks for this! just uploaded [06:11] jbicha: ah, sorry, no; I merged your g-s-d branch [06:11] I didn't see a g-p-m MP yet [06:12] pitti: https://code.launchpad.net/~jbicha/ubuntu/oneiric/gnome-power-manager/gnome-power-manager-3.1.2/+merge/68165 [06:13] ok, time to reboot after a week of updates, crossing fingers :) [06:13] ah, /me uncommits his 13-energy-star.patch patch cleanup then and merges your's [06:13] didrocks: you missed all the /run fun! [06:14] pitti: I followed it by emails, it was enough of "fun" :-) [06:14] pitti: oh, good thing I said something [06:22] jbicha: I'm surprised that g-p-m still has the icon-policy key [06:22] I thought that moved into g-s-d now [06:22] respectively indicator-power [06:23] it probably should be there [06:23] anyway, looks good, thanks! [06:23] by the way, these override settings don't seem to work for me [06:25] none of them appear to change the defaults or even get applied at all [06:25] jbicha: in g-s-d you mean? [06:26] pitti: I mean not one of the gschema.override's installed here seem to actually do anythinig [06:36] pitti: oh, never mind, it's just dconf-editor that's broken [06:36] oh? it usually seems to work for me, including picking up overrides [06:52] waow, the CD is really oversized. What happened? [06:54] mvo: hey how are you? :) [06:54] hey didrocks! I'm good, how are you? [06:54] mvo: I'm fine, thanks! I managed to do some good progress on oneconf during my RMLL trip :) [06:55] mvo: not on #oneconf? [06:57] didrocks: apparently nobody heard my plea for rebuilding ubuntu-meta on Friday, doing now :) [06:58] didrocks: we added ibus-sunpinyin, but it's too big [06:58] originally I thought that ibus-pinyin was 15 MB (that's what apt-get install does), but due to our clever seeds of preferring the android db, it's only 1.5 [06:58] so rebuilding meta will get it down by some 17 MB [06:58] the bad news is that now we once again have no plan how to get rid of the remaining 13 MB [06:59] pitti: heh, that's worth an ubuntu-meta rebuild, indeed :) [06:59] hum… no more langpack to remove? [06:59] the original deal for tbird was to build a Chinese image and dropping ibus-pinyin and chinese langpack [06:59] we can still remove the Chinese langpack then [06:59] which will give us back 6.5 MB [06:59] but that will enable to get a chinese image of 703 MB? still [07:00] and chrisccoulson said that the recent 3 MB firefox growth should be fixed in the next build [07:01] ah, need to re-promote ibus-pinyin [07:01] If cairo drops its egl backend then libegl and friends will drop off the CD, too. [07:01] oh, will it? [07:02] I guess it still causes memory hogs with nvidia? [07:02] Yes, yes it does. [07:02] Unless someone does the work to make it dlopen stuff, it's highly likely we'll just disable the backend again. [07:02] (This was Seb's plan) [07:03] That'd drop a couple of meg. [07:04] glibmm will soon be dropped from compiz and unity, but there is still gparted depending on it, I think [07:05] And if we make it dlopen stuff, then we can drop egl from the cd anyway. [07:05] * RAOF → shop. [07:07] didrocks: and gnome-system-monitor? [07:08] pitti: oh, even the gnome3 one? I didn't check… ok, no hope on that direction then [07:08] didrocks: is there a gnome 3 one now? [07:08] didrocks: a few weeks ago, cjwatson made a first pass at porting to gtkmm3.0 [07:08] didrocks: but it would only move from gtkmm 2.4 to 3.0, so no real difference [07:09] pitti: not sure, didn't follow that TBH. So you seem more up to date than I there :) [07:13] pitti: I was working on packaging a newer gnome-shell but it doesn't have a ~ubuntu-desktop branch [07:14] I guess it's in universe, it that why? [07:14] mostly, yes [07:15] jbicha: main reason is that we would like it to stay in sync :) [07:15] Morning desktopers! [07:15] jbicha: seems our only delta is the gir1.2-json-1.0 rename, which hopefully happened in Debian as well? [07:15] hey Sweetshark, how are you? [07:16] pitti: yeah it looks like the only change now is that we have different default favorite-apps [07:16] jbicha: yes, it did [07:16] jbicha: i. e. 3.0.2-1ubuntu1 was only a temporary delta, and the first one ever [07:17] good morning Sweetshark [07:17] pitti: are we going to keep the favorite-apps delta though? [07:18] jbicha: I don't see that delta yet, is it in a merge proposal? [07:18] * pitti is looking at apt-get changelog gnome-shell [07:19] * pitti checks if the json rename is in svn [07:19] it's not, I'll commit it there [07:20] pitti: it's 01_favorite_apps.patch and it is part of 3.0.2-1ubuntu1 [07:20] jbicha: that's in Debian [07:21] it changes mozilla-firefox.desktop to epiphany.desktop [07:21] except Debian does it wrong ;-) [07:22] the Ubuntu version has firefox, libreoffice, banshee, & perhaps should have thunderbird [07:23] oh you're right, Ubuntu didn't diverge [07:23] but should we? [07:23] * didrocks thinks we shouldn't diverge apart if someone is willing to maintain gnome-shell in ubuntu [07:24] if we can change that only if dpkg-vendor --is ubuntu, sure [07:24] * debian/control.in, debian/rules: When building on Ubuntu, add a dependency [07:24] to gnome-icon-theme-full. [07:24] ^ I did that for our gnome-icon-theme split [07:24] ok, is Debian going to have Gnome 3.2 ready for Oneiric's release? [07:24] jbicha: that's the plan [07:24] erm, sorry, misread [07:25] jbicha: no, Debian will stay at 3.0 until 3.2 is released [07:25] we are packaging 3.2 for Ubuntu right now [07:25] micahg: committed your gnome-shell json dependency fix to debian svn, thanks [07:26] hum, my usb key isn't automounted, dmesg tells me all is fine, should I check udev? [07:26] didrocks: do you see it in nautilus? [07:27] no, I don't [07:27] didrocks: in gvfs-mount -li ? [07:29] pitti: no, I just have my optical drive listed [07:31] didrocks: do you see it in udisks --dump? [07:31] udisks --enumerate-device-files might be easier to read [07:32] pitti: yeah, it's in! [07:33] ok, so something in gvfs [07:34] * didrocks looks at recent uploads [07:34] didrocks: can you remove it again, start "gvfs-mount -oi", and plug it in? [07:34] let's see what happens [07:34] sure [07:35] pitti: seems that it doesn't notice anything :/ [07:35] didrocks: but you do see it in "udisks --monitor-detail" if you unplug/replug/ [07:35] ? [07:36] pitti: yeah, it's listed there [07:37] but nothing in gvfs-mount -oi meanwhile [07:37] pitti: I got mails about translation template imports for lo-build in natty, did you upload 3.3.3 to natty-proposed while I was ill? [07:37] is gvfs-gdu-volume-monitor running? [07:37] Sweetshark: it got accepted into natty-proposed on Friday, yes [07:37] pitti: it even seems to be happily running [07:37] Sweetshark: feeling better now? [07:37] pitti: yes, Ready to conquer the world! [07:38] should I kill it and run it by hand? [07:38] didrocks: could you perhaps attach an strace -fvv to it and do the replug exercise again? [07:38] pitti: sure, trying [07:40] pitti: 3.4.1 for oneiric getting its finishing touches (some misbehaving script for creating libebook deps) and I will leave out the mono stuff for the first try. [07:46] pitti: sorry, had to do it twice, because gvfs-gdu-volume-monitor needs to be ran in /. Here are the trace with connecting, disconnecting: http://paste.ubuntu.com/646280/ [07:47] argh, an -s1024 would be helpful there, too, to be able to read the dbus messages [07:47] but anyway, looking [07:48] can do :) [07:48] bah, why does it read /etc/fstab a million times [07:48] didrocks: but the device is not in there, is it? [07:49] pitti: no, it's not listed there, I just checked again [07:50] last time I tried this ubuntu usb stick it worked, I should probably try to swap to another one, just in case… [07:50] didrocks: you can try udisks --mount /dev/sdb1 [07:50] if that works, the stick is okay [07:51] didrocks: can you give me the udisks --dump output while it's plugged in? [07:51] pitti: indeed, udisks --mount /dev/sdb1 works [07:52] pitti: http://paste.ubuntu.com/646284/ [07:53] uh, that looks screwed up [07:53] hum, I can see another usb stick randomely picked [07:54] pitti: what seems screwed? [07:54] looks like both /dev/sdb and sdb1 claim to be a mountable file system [07:54] (iso 9660) [07:54] but it should still automount sdb1 [07:54] oh right, nice [07:54] hence the manual mount working [07:55] pitti: I've got the Ubuntu favorites patch for Debian, can I email it to you? [07:56] jbicha: oh, great! sure [07:56] hah! take that, gsettings-data-convert [07:56] no more crashing [07:56] pitti: probably usb-creator wasn't in a good shape last time I created the key. But anyway, as you told, it should pick one part to mount anyway [07:56] pitti: nice \o/ [07:57] pitti: thanks a bunch for helping debugging and the explanation ;) [07:57] good morning everyone [07:57] hey chrisccoulson! [07:57] hi didrocks, how are you? [07:58] chrisccoulson: I'm fine, thanks, good and long week-end ;) and you? [07:58] I can see a thunderbird with nice notification, thanks! :) [07:58] hey chrisccoulson [07:59] didrocks: how big is this thing? [07:59] yeah, i'm good too thanks. i had a long-ish weekend, and went to stay with some friends. [07:59] hi pitti, how are you [07:59] ? [07:59] didrocks: perhaps I can replicate it here if you dump the first MB to a file, and I put it on an USB stick of mine with an usb-creator output on it? [07:59] chrisccoulson: I'm great, thanks! [08:00] * Sweetshark dances. 3.4.1 finished (without mono, but still). [08:00] pitti: it's an ubuntu iso, nothing more. Can do a dd for the first MB, sure [08:01] pitti - it seems that the PGO build failure on friday is i386-specific (which is why it worked locally) [08:01] that's a bit weird still :/ [08:01] didrocks: right, but with the specialty of the partition table/first blockes being messed up [08:03] pitti: dd if=/dev/sdb of=messedupkey count=2048 sounds good to you? [08:03] chrisccoulson: is that a compiler bug, or does firefox have platform specific code? [08:03] didrocks: I"m not soure what the default bs is; perhaps try bs=1MB count=1 ? [08:03] pitti - the code which fails isn't platform specific, so i'm not sure yet [08:04] that can work too :-) [08:05] pitti: here we go: http://people.canonical.com/~didrocks/messedupkey [08:05] didrocks: cheers; just uploaded gconf for the session start crash, so I can now look into that [08:06] pitti: excellent, thanks a lot :) [08:07] jbicha: ah, I'm not a big fan of ubuntu.series -- it has to be kept up to date [08:07] it's not applied in addition to series [08:07] jbicha: perhaps we can use sed in debian/rules instead of a patch? [08:10] pitti: hmm, I think I sorta understand what you're saying but I haven't done anything like that before [08:10] didrocks: wait, your's actually says iso9660? [08:10] I haven't done ubuntu.series before either but I was able to read the man page for that [08:10] didrocks: usb-creator creates and writes FAT [08:11] didrocks: perhaps you tried dd'ing an .iso first to sdb then to sdb1? [08:11] didrocks: as cjwatson announced recently we should now have hybrid images, so that would actually work [08:11] pitti: oh oh… you are probably right, I'm remember doing that some weeks ago [08:11] for testing it [08:11] didrocks: does it work? [08:12] didrocks: just trying to recreate your steps to get an USB stick which is roughly the same [08:12] pitti: yeah, it clearly does :) [08:12] nice [08:13] pitti: so blame the user :-) me in that case! anyway, trying to mount the first part in case there are both would still be a great addition :) [08:13] didrocks: yes, I know [08:13] hmmm, it's a shame that the 10.04.3 ISO won't have a nice shiny new firefox version on it ;) [08:13] didrocks: I'm not arguing it's not a bug, just trying to recreate your broken usb stick [08:14] pitti: right, you totally right, quite crazy I forgot that, sorry :/ [08:15] pitti - oh, you fixed the gconf crash? i was going to look at that this morning :) [08:15] chrisccoulson: yes [08:15] thanks [08:15] no worries ;) [08:15] slightly fewer apport spam dialogs at session start [08:16] the "system crashes" ones are annoying enough [08:20] chrisccoulson: do you have the "Sharing Options" by right click in a nautilus folder? [08:20] didrocks, no, it doesn't look like it [08:21] chrisccoulson: seems I'm not crazy then :) [08:25] didrocks: ah, no need for your image even -- merely dd'ing the iso on an USB stick reproduces that very bug [08:25] hi! [08:26] bonjour seb128! had a nice long weekend? [08:26] pitti: oh nice :) [08:26] hey seb128! [08:26] pitti, hey, excellent thanks! [08:26] hey didrocks, how was your week off? [08:26] seb128: it wasn't a week off ;) [08:26] rrrrright ;-) [08:26] but was nice [08:27] seb128: even some people forced me to fix unity-2d at 02:00 AM :p [08:27] next you will tell me GUADEC is an hard work week as well :-p [08:27] lol [08:27] didn't say it's an hard work, but people was disappointed to not see you so close to your home :) [08:29] lol, you told them that some people have work to do? ;-) [08:30] seb128: it wasn't a slacking week… but oh well, no time to argue, I have work to follow on [08:30] didrocks, right, have a week off I bet you have ;-) [08:30] … [08:30] ok, I stop trolling you there :p [08:31] chrisccoulson: btw, I have now the answer on your thunderbird not getting new emails on some folder before clicking on them: the indicator doesn't show them as well [08:34] seb128: fwiw, I think it's important from a community perspective to have developers going to events like rmll [08:35] hey vuntz [08:36] vuntz, I'm sure didrocks got you to say that in exchange of something ;-) [08:37] vuntz, but right, I don't disagree, it's just that cycles have a limited number of days and hours and it's hard to get work done and to attend all the conferences around as well [08:37] oh, I tried to get him to break some stuff, but that didn't work [08:39] vuntz, did you manage to second review the wnck patch? ;-) [08:40] I'm waiting for some angel telling me what's the best way to name the API [08:40] seb128: didier told me that he relaxed a lot and did almost no work at rmll [08:40] seb128: i wouldn't have high expectations of what he accomplished while there [08:40] desrt: :-) [08:41] desrt, right, next he's going to tell people that GUADEC is a lot of work, I tell you ;-) [08:41] GUADEC is a lot of work [08:41] ...for ones liver [08:42] seb128: did I tell that? but ok, meeting the openshot core dev, building the community, taking the time to explain why unity and having people changing their opinion is not that important… [08:43] didrocks, let's stop that, I was mostly playing a bit with you, no worry ;-) [08:43] hug! [08:44] pitti, oh, you fixed that gconf segfault, great [08:46] didrocks, oh, happy piloting btw ;-) [08:46] seb128: I piloted the week before RMLL to swap with last Friday [08:46] but thanks for the reminder nevertheless [08:46] oh ok, I'm catching up on calendar and stuff since I was off friday [08:47] seb128: yay less apport spam ;) [08:48] nice, the new gwibber landed, I can clean some extra gtk2 gconf lines [08:48] but robert_ancell still didn't fix lightdm :-( [08:48] hum, I wonder if he noticed it was broken, I will drop him an email [08:50] What's broken about lightdm? [08:50] * RAOF is pretty sure he's got the latest packages, and everything *seems* to be working when he booted today… [08:51] the example greeter (which is default) segfault on protected userdirectories [08:51] it seems to not like not being able to read the .dmrc [08:51] Ah. And neither of those statements apply to my system, which would be why I haven't noticed :) [08:51] I use ecryptfs for my home, but the unity greeter [08:52] didrocks, oh, interesting. thanks [08:52] the unity greeter doesn't have the issue [08:52] hey chrisccoulson, how are you? [08:52] seb128, yeah, good thanks. how are you? [08:52] I'm fine thanks [08:53] ideally we will get the unity greeter mir-reviewed and promoted soon but still would be nice to fix the default greeter [09:00] pitti, btw did you see my py3cairo preinst comment before the w.e? [09:00] well middle of previous week rather, before my w.e [09:00] seb128: I did [09:00] seb128: I reuploaded to NEW with the preinst removed [09:00] ok, will re-review it [09:01] merci [09:01] it seems fine out of that bit [09:01] de rien ;-) [09:01] seb128: I also sent it to the Debian pycairo maintainers, but haven't heard back from them yet [09:01] yeah, he's usually not very active or responsive [09:01] seb128: but I found a way to build python3-gobject without it, so I could do this large change in Debian [09:01] and we can enable the py3cairo bits once py3cairo lands in debian [09:05] didrocks, do you use a local filter in thunderbird to move your mail between folders btw, or do you do filtering elsewhere? [09:06] chrisccoulson: filtering is done server side, nothing local [09:06] didrocks, aha [09:06] i guess that might be why [09:06] i think thunderbird checks those folders lazily [09:06] oh… [09:06] i'm not 100% sure on that though [09:07] some seems to be refresh quite frequently, but it's a huge blocker though :/ [09:08] didrocks, perhaps try right-clicking on the folder and selecting "When getting new messages for this account, always check this folder" (not sure what that translates to in french though) [09:08] (in the folder properties) [09:08] didrocks, i don't see the issue because i do my filtering locally [09:08] chrisccoulson: trying and will tell you [09:08] thanks [09:10] yw :) [09:12] hi [09:15] hey rodrigo_ [09:17] morning rodrigo_ [09:17] hey rodrigo_, how are you [09:18] hi seb128, pitti, didrocks, how are you? [09:18] rodrigo_: I'm fine, thanks, and you? [09:18] rodrigo_: quite fine, thanks! [09:18] how about yourself? [09:18] I'm a bit tired since I went to sleep quite late last night, but fine otherwise :) [09:19] didrocks, seb128: enjoyed your 14th july? :) [09:19] yes, very much! [09:20] nice slacking w.e ;-) [09:20] :) [09:20] nice fireworks in Lyon in the evening, I worked in the train and catchup on emails once back but trade with the 15th [09:25] but no head cutting in the celebrations, right? :) [09:28] no ;-) [09:29] pitti, did you just restart the retracers today? [09:29] seb128: yes, fixed up the broken amd64 chroot [09:29] seb128: well, "just" - that was maybe 1.5 hours ago [09:30] ok, good, I was wondering if they were catching up on backlog or if we get a stream of new bugs ;-) [09:30] reading through my bug email box it's retracer spammed [09:30] but if that's backlog catching up that's ok [09:30] pitti, thanks ;-) [09:36] does the new me-menu work for anyone? [09:40] it doesn't for me, and it breaks stacking of other menus [09:41] same here [09:41] and it pointlessly shows my name [09:42] +1 and my name is fairly long. it uses 15% of panel width :/ [09:42] it should probably show your status ("Away") instead, which would make more sense [09:42] or at least the user name, not the real name [09:42] didrocks, does the skip bar flag you want to use in nautilus means also it will not be in the alt-tab list? [09:43] seb128: right, for John, it's acceptable [09:43] didrocks, what about non unity sessions? [09:44] seb128: don't know, I think we don't have a nice story there [09:44] hum [09:44] I raised it during the discussion for what it worth [09:44] didrocks, can we have those discussions in public or better on the bug next time? [09:44] seb128: sure [09:44] I'm not sure I agree with that change, having dialogs not in alt-tab really sucks [09:45] especially when you are not under unity and you don't have a launcher [09:45] didrocks, thanks [09:45] not being able to open nautilus during a copy isn't nice as well [09:47] well you can middle click the icon ;-) [09:47] not really well advertized :-) but the real solution as told, will be to do some heavy tab-related work in bamf [09:48] right, let's see, I guess they are not going to do that refactoring during the lts cycle either [09:48] so let's see how it goes in practice [09:48] we should make the copy dialog a standalone process so it has its own icon :p [09:49] integrated into compiz? :-) [09:51] why not ;-) [09:57] what is the current convention for dbus services "CamelCaseFunctions" or "pep8_style_lowercase" ? or something else? [09:58] hey seb128, welcome back, apt with hold-back-for-things-that-would-break-recommends has lnaded (we talked about it during the rally) [09:58] mvo, hey, great! [09:58] (dbus question is regarding the new software-propoerteies frontend/backend split) [09:59] mvo, will you get ride of gksudo there? [09:59] dunno about if there is a naming convention, will let pitti or others reply ;-) [10:00] seb128: yeah, no more gksu there [10:00] seb128: plus better reflection of reality [10:00] \o/ [10:00] :) [10:00] *but* lucnhtime ;) [10:24] didrocks: hm, unity-support-test says "no" in a guest session for me, in particilar "not software rendered: no" [10:25] pitti: whereas it says "yes" in the main session? [10:25] right [10:25] * pitti tries another user account [10:25] can be a Xauth capability with apparmor if it's only the guest session [10:25] some similar issues happened twice in the natty cycle [10:26] whoops, I get the example-greeter crash [10:26] (even though I run unity-greeter by default) [10:27] pitti, see ;-) [10:28] pitti, is that specific to guest session or do you get it with a test user? [10:28] seb128: guest session works [10:28] freshly created test user crashes [10:28] weird [10:28] pitti: you can try glxinfo | grep rendering btw, should give the same result (same X call) than the support test [10:28] and unity-2d doesn't respect my /usr/share/gconf/ubuntu-2d/default/zz_ubuntu-defaults-test overrides any more [10:29] didrocks: already tried that; it does say "direct rendering: Yes" in guest session [10:29] hum, weird, we use the same flag, let me check [10:29] pitti: right, the overrides are fix is "fix committed" [10:30] didrocks: you mean it doesn't read gconf any more right now? [10:30] so guest doesn't work with 3d; switching to test user is broken; unity-2d doesn't read gconf settings [10:30] pitti: well, depends, if it fallbacks, it won't until we set the detection in lightdm itself. If you choose the session, the trigger wasn't… triggered [10:31] how am I supposed to test ubuntu-defaults-builder now? :-) [10:31] pitti, kvm for the win? ;-) [10:31] didrocks: oh, you mean I could explicitly set the default session to unity-2d? [10:31] pitti: right, the other has never worked for default settings [10:32] (fallbacking) [10:32] didrocks: hm, as it happened /usr/share/xsessions/default.desktop already pointed to ubuntu-2d.desktop [10:32] * pitti pokes this harder [10:32] what is /usr/share/xsessions/default.desktop ? [10:32] I should miss some updates, I don't have it there [10:33] I don't know where it's coming from [10:33] * pitti tries with lightdm.conf [10:34] oh! lightdm doesn't tear down the guest account [10:35] didrocks: so you were saying if unity-2d is selected explicitly, the Xsession.d script should add some $XDG_something to make it pick up the gconf defaults? [10:35] didrocks, it's not a distro package, it's probably an hacking leftover for default session in gdm or something [10:35] didrocks, I've no such file installed either [10:35] pitti: right [10:35] didrocks: now, doesn't seem to work :( [10:36]