[01:06] achiang: the problem is I do need to do the first 24k updates, and that still means nearly an hour at 90ms per update I think? [01:08] hm... [01:08] cyphermox: well, easier to repro than my 24 hours ;) [01:09] it is ;) [01:09] and so far it seems pretty foolproof [01:11] good to have a reproducer. makes me feel like you'll solve it. :) [01:12] still need to dive yet again in the code to find out where things went wrong, but yeah the furture is getting brigther [01:12] * cyphermox takes that opportunity to make a new debian vm [01:16] not bad, already reaching 10k [01:16] I should figure out a way to slow it down via some file or whatever [01:25] Stupid facebook API documentation is *terrible*. Imagine trying to read an API documentation that listed only methods and return values, no call signatures. There is literally zero information about the supported parameters to these REST API methods... I am literally reduced to using the API explorer and guessing call arguments until i stumble on the effects that I want. [01:25] Curse you zuckerberg!!! [01:42] achiang: it's not an overflow, barely 20k and already broke === Ursinha-afk is now known as Ursinha === robert_ancell is now known as robert_ancell|di [04:55] Good morning [06:31] good morning [06:33] robru: no open source clients you could take apart? :p [06:35] bonjour didrocks [06:40] guten morgen pitti, wie geht's? [06:41] didrocks: much better now that I found out (and fixed) why people get boot breakage with raring-proposed :) [06:41] :) [06:42] other than that: Je veux revenier au printemps!!! [06:42] I again biked through snow last night [06:45] :) [06:45] * didrocks sees blue sky and sun here [06:45] here too again, but it's still cold [06:45] yesterday, had a very little bit of snow (went from 10° to 2°) [06:45] but the temperatures are up again \o/ === robert_ancell|di is now known as robert_ancell === tkamppeter__ is now known as tkamppeter [07:44] good morning [07:45] hey chrisccoulson! [07:46] chrisccoulson: congrats :) [07:46] good morning [07:46] thanks :) [07:46] how are you? [07:46] chrisccoulson: can I mark my thunderbird hangup as a security issue? ;) [07:46] hah ;) [07:46] lol [07:46] chrisccoulson: I'm good, thanks! Hoping today will be less crazy than the previous days :) [07:46] and you? [07:46] hey larsu! [07:46] salut jibel [07:46] good morning everyone :) [07:46] how is London? [07:47] cold! [07:47] bonjour didrocks [07:47] but sunny this morning [07:48] :) [08:15] pitti: hey! not sure if you can bump ppa builds (virtualized ones). i386 built, but it would be coll to have amd64 building quickly for those: https://launchpad.net/~ubuntu-unity/+archive/experimental-prevalidation/+build/4369127 [08:16] pitti: there will be unity in the near future, but I prefer to wait libunity building on amd64 first as otherwise, it will just dep-wait [08:16] didrocks: you, je peux [08:16] err, "oui" [08:16] how on earth did I type "you"? [08:18] pitti: same group of letters :) [08:18] pitti: Danke Schön! [08:18] de rien, mon ami! [08:19] will bother you again on Unity later [08:19] pitti: btw, do you know how oftenly dep-wait are retried on a ppa? [08:19] pitti: if they are at all? [08:19] didrocks: distro builders retry on every publisher update AFAIK; no idea about PPAs I'm afraid [08:20] c'est un question de infinity [08:20] pitti: yeah, I'm on the ppa one, I'll retry manually the ones I'm really interesting in [08:20] pitti: let's try it in French with infinity :-) [08:21] didrocks: il l'aime! [08:21] :) [08:21] speaking of unity, how do I tell it to use as little memory as possible? for bug 1121715 [08:21] Launchpad bug 1121715 in xserver-xorg-video-nouveau (Ubuntu) "Unity corruption with nouveau driver on GeForce 7025" [Undecided,New] https://launchpad.net/bugs/1121715 [08:22] mlankhorst: would be a question for upstream on #ubuntu-unity [08:22] mlankhorst: I would say disable plugins [08:28] didrocks: it just occurred to me -- is there any pronounciation difference between "Vincent" and "25" (vingt-cinq)? [08:29] just the "k" at the end? [08:29] pitti: no, it's a little bit different [08:29] like vincent would be "20 100" [08:29] (vingt cent) [08:30] aah [08:30] cinq is more like "un" (1) [08:30] so "un" "en" are quite similar, but the pronouciation is different [08:30] ok, you gotta teach me the difference in Oakland or so :) [08:30] pitti: will do for sure! :-) [08:31] while running in the morning exercice :) [08:32] c'est un bon occasion pour le comptage :) [08:32] "1001, 1002, 1003..." [08:32] héhé, tu veux compter le nombre de pas? c'est un peu rapide ;) [08:33] * pitti peut compter à 100 seulement [08:34] ça fait un peu court comme course :-) [08:34] on comptera les maisons! [08:34] haha [08:34] didrocks: I'll just wrap around after 99 :) [08:35] ;) [08:38] on a more serious note, I'll have a go at adding logind support to indicator-session [08:38] sweet! [08:38] it's talking to gnome-session for shutdown/reboot also seems broken [08:38] calling gnome-session's shutdown through d-feet works fine, but from the indicator it's working [08:38] err, s/working/broken/ [08:38] urgh [08:38] even with latest unity? [08:38] because unity has the shutdown dialog now [08:38] well, it falls back to ConsoleKit, so I guess we just never noticed [08:38] was working when trying it [08:38] ah [08:38] hm, not here [08:38] well, I'll debug it; I guess nobody else is working on that right now? larsu? [08:38] I don't think so… [08:38] pitti: mind bumpint the priority for those, now that libunity is published: https://launchpad.net/~ubuntu-unity/+archive/experimental-prevalidation/+build/4369085 [08:38] https://launchpad.net/~ubuntu-unity/+archive/experimental-prevalidation/+build/4369082 [08:39] and https://launchpad.net/~ubuntu-unity/+archive/experimental-prevalidation/+build/4369067 [08:39] didrocks: comment tu dis "done!" à français ? [08:40] "dis-tu" [08:40] pitti: "fait" [08:40] or rather: [08:40] "fait !" [08:40] oui, c'est simple [08:40] (notice the space ;)) [08:40] didrocks: fait ! [08:40] pitti: merci beaucoup ;) [08:40] didrocks: yeah, I have a really hard time teaching myself to use improper spacing before punctuation :) [08:41] pitti: rohhh "improper" :-) [08:41] hey desktopers [08:41] lut didrocks pitti [08:41] salut seb128 [08:41] * pitti donne une accolate à didrocks [08:41] ... et seb128 [08:41] * seb128 hugs pitti back [08:41] * didrocks donne une accolade à pitti ;) [08:41] excusez-moi, "accolade" (typo) [08:42] seb128: comment vas-tu ce matin? [08:42] seb128: comment vas-tu ce matin ? [08:42] grrr improper spacing! [08:42] pitti, je veux bien, mais j'en ai assez du froid, je veux le printemps ! [08:42] pitti, et toi ? [08:42] seb128: mon aussi ! [08:43] * didrocks scrutinize the correct use of spacing finally on this channel :) [08:43] seb128: mais il ne neige plus ici [08:43] pitti, ici non plus, mais c'est gelé avec de la neige au sol [08:44] * pitti grabs dictionary for gelé [08:44] ah [08:45] nous avons mangé de la glace semaine dernière ! [08:45] c'était presque le printemps : -( === jibel_ is now known as jibel [08:49] make[3]: Entering directory `/home/martin/upstream/indicator-session/tests' [08:49] make[3]: *** No rule to make target `/usr/src/gtest/gtest-all.cc', needed by `gtest-all.o'. Stop. [08:49] HALP! [08:49] I do have libgtest-dev [08:49] larsu: any idea? [08:50] I have a /usr/src/gtest/src/gtest-all.cc [08:51] hm, I ran autogen again, works now; sorry for the noise === duflu_ is now known as duflu [09:06] pitti: I'm letting charles know anyway ;) [09:06] pitti, we could fix this but I'm not sure it's an interesting thing to fix: [09:06] pitti: I'm going to be landing a Large rewrite of the tests in i-session in the next couple of days [09:07] if the problem persists after that then it would be more interesting [09:09] morning! [09:11] hey Laney [09:11] hey pitti, how are you? [09:12] Laney: quite fine, thanks! much better after fixing broken boot this morning :) [09:12] this transition seems to be keeping you busy :-) [09:12] heh, you bet [09:12] Laney: Lennart was quite right with "we'll have pain" [09:13] I called him last night to discuss the sd_booted() API mess, and we got some agreement [09:13] new API? [09:13] Laney: but the boot fix was due to an ancient hack in libudev which I didn't port, and had long forgotten about [09:14] what does he think of logind being used standalone? was that an intended usecase frmo the start? [09:14] no, not at all (and it still isn't) [09:14] but he agreed to changing sd_booted() to actually mean "systemd init", and giving us a function "shoudl I talk to logind" [09:15] then we "just" need to fix the 20-or-so packages in Debian/Ubuntu/upstream which use sd_booted() and mean "logind" [09:15] * Laney nods [09:15] but I reall want some help there [09:16] well if you mail about what you need I'm sure some people (myself included) can help [09:16] yeah, once we have a final settlement between upstream, Debian, and us how this should look like [09:17] see my ping in #-devel last night, BTW? Broken audio and brightness adjustment with the PPA pkgs [09:17] Laney: it got drowned in the OMGboot messages, sorry [09:18] Laney: getfacl /dev/snd/pcmC0D0c ? [09:18] do you have an ACL there? [09:18] I ppa-purged it already, sorry [09:18] needed to watch a youtube video to tell me how to cook dinner /o\ [09:18] ah, ok [09:19] I would upgrade again but I don't see a systemd package in there any more? [09:19] Laney: right, I removed it because that's in raring-proposed now [09:19] (with some fixes) [09:19] I did a full dist-upgrade to raring-proposed this morning, works fine here [09:19] (that includes the libudev1 transition) [09:20] it will kill me with half done haskell, but perhaps I can hold that back [09:20] to the aptatron [09:20] that reminds me, pulseaudio needs to be fixed harder, apparently it has a hardcoded libudev0 dpe [09:20] oh [09:20] Laney: or just grab the debs from LP? [09:20] all of libudev1? sounds annoying [09:21] Laney: no, just systemd-services and libudev1, and the libsystemd* libs [09:22] huh, no - apparently I have already transitioned all of the ghc bits that I have installed :-) [09:22] dist-upgrade looks fine [09:26] * Laney reboot [09:28] yeah, seems to work still [09:28] ooh [09:29] didrocks, larsu, charles: so is indicator-session actually being used for shutdown? I see ConsoleKit reboot/logout stuff in unity [09:29] but these also try gnome-session first and fallback to CK [09:29] Trevinho_: ^^ [09:30] pitti, it will call the unity shutdown dialog; but if that's not present, indicator-session has a small utility that calls ConsoleKit itself [09:30] right, but gtk-logout-helper -r also works fine (talking to gnome-session) [09:31] so I wonder why calling gnome-session Reboot() works, gtk-logout-helper works, but when I click reboot in the indicator menu it just brings me back to lightdm [09:31] perhaps it tells the session to shutdown first, so that gnome-session's D-BUS API already went away? [09:32] and thus it falls back to the (nonexisting) CK API? [09:32] that would explain why I do get logged out, but not rebooted [09:33] pitti, I'm in the middle of a blueprint planning right now and will have to get back to you in ~1 hr [09:33] ack, thanks [09:34] hi all. any idea how I can make ubuntu *not* flash the problem report dialog *even* if something crashes? This system is for automated use, and I don't want those dialogs [09:34] RoyK: uninstall apport, or set "enabled=0" in /etc/default/apport [09:36] thanks [09:40] didrocks: can we get bug 1155021 into the attention pipeline somehow? [09:40] Launchpad bug 1155021 in unity (Ubuntu) "Add logind support (ConsoleKit is being deprecated)" [Undecided,New] https://launchpad.net/bugs/1155021 [09:41] pitti: I can talk about it, but with 100 scopes project urgency, I doubt it will be for raring [09:41] pitti, ping Trevinho_ about it [09:41] he just wrote that code [09:41] seb128: I've already pinged him :) [09:41] k [09:41] 10:29:35 didrocks | Trevinho_: ^^ [09:42] it's quite simple code, but I'd like to at least discuss with upstream about how they want this conditionalized [09:42] i. e. configure option, etc. [09:42] the actual code to call logind is a simple copy&paste exercise [09:43] pitti: I'll keep the tab opened and discuss about it next meeting if Trevinho_ doesn't answer first (next Tuesday) [09:45] actually, I'd really like to find out what's wrong with calling gnome-session, that just DTRT [09:46] and would avoid duplicating this over indicator-session and unity [09:47] indeed, makes sense, let's wait for him to be online [09:48] Laney: still no ACLs on /dev/snd/* ? [09:50] pitti: looks fine (and works) http://paste.ubuntu.com/5613166/ [09:51] so either fixed or it was logind ;-) [09:51] when it was broken I had nothing other than Dummy Output in the sound capplet [09:54] Laney, oh btw, you added an upstart job to indicator-weather, does it mean it's ran twice now (I think you didn't desactivate the autostart for it)? [09:56] no, it just doesn't launch the second time [09:57] you mean? [09:57] indicator-weather is single instance and will exit if one is running? [09:57] indicator-weather is already running; doing nothing [09:57] or so [09:57] right [09:57] so we still waste cycles to start something that then exit [09:57] not a big deal but it's something we should consider when we start adding upstart jobs [09:58] yeah [09:58] my idea was to drop Unity from the autostart .desktop when we add the jobs [09:58] it's hard at this point because they're opt-in [09:58] right [09:58] so you have to say "is this session running under upstart? then ignore the autostart file" [10:00] yeah, it's a bit tricky [10:00] one easy way would be to say "unity is upstart based" [10:00] then we can start migrating stuff by dropping unity from the .desktop when we add the upstart jobs [10:00] it's a bit tedious and not really flexible though [10:01] it's still not on by default even for that [10:01] you can test $UPSTART_SESSION being nonempty [10:02] test -n "$UPSTART_SESSION" && old_command or so [10:02] right [10:03] or ... [10:04] is that directory specified by an $XDG_... env var? [10:04] we could prepend to that to shadow it [10:05] ah, $XDG_CONFIG_HOME is a single directory so that doesn't work [10:07] I would just go for "make unity an upstart user session" and once that's done, add the jobs/drop Unity from the .desktop targets through the components [10:07] that's easy enough for Unity [10:08] but doesn't really work out that well for other desktop if they don't do it the same sort of transition [10:13] pitti: are you really sure you did push the trigger for https://launchpad.net/~ubuntu-unity/+archive/experimental-prevalidation/+build/4369085? it seems we still have the default score :) [10:13] didrocks: done now; I may have grabbed the wrong url, sorry [10:13] pitti: no worry, thanks! === vrruiz_ is now known as rvr_ [10:25] Laney: oh, I know what could be the reason of temporary ACL failure [10:27] Laney: I didn't disable udev-acl in the PPA yet, so both logind and udev-acl will fiddle with ACLs; the rule in /lib/udev/rules.d/70-udev-acl.rules which is supposed to prevent that doesn't help [10:32] ah yes, makes sense that logind will have to do this [10:35] Laney: udev_175-0ubuntu21_source.changes uploaded with a fix [10:36] I'll completely drop udev-acl once we land logind as a default, but until then this will do nicely to get out of the way [10:38] I suppose that explains brightness too [10:38] merci! [10:41] pitti, didrocks, seb128: so I'm back === Trevinho_ is now known as Trevinho [10:42] Trevinho, hey [10:42] hi seb128, thanks for merging the nautilus patches [10:42] Trevinho, yw, does it need an unity fix as well? [10:43] pitti: about ConsoleKit in unity, I just implemented that as a fallback in the remote case that gnome-session fails [10:43] seb128: for having the icons illumination, yes [10:43] Trevinho, click on an usb stick icon in the unity launcher still opens several instance of nautilus rather than focussing the existing one [10:45] seb128: it should open a new one, but clicking again on it it re-uses the same view [10:45] Trevinho, not here, it keeps opening new views [10:45] seb128: o_O it works here [10:46] seb128: it re-open a new one if you're in a sub dir of the opened one [10:46] well, I start a fresh session, plug an usb stick, click 5 times on the unity icon [10:46] I get 5 nautilus views on /media/ [10:46] seb128: we can in case fix this, but only for ubuntu since upstream can't accept a such behavior for everything [10:47] o_O [10:47] doing "nautilus /media/" from a command line doesn't open new ones [10:47] not sure what the launcher is calling [10:47] seb128: ah, wait... let me think [10:48] seb128: mh, I've just pushed to unity some patches for this (but mostly related to the focus) [10:48] and it works here, so... let me check how it would work without them [10:50] seb128: it shouldn't btw, since unity is calling the same path as you'd do by command line [10:50] Trevinho: that's what I figured, but the g-session call doesn't seem to work [10:50] seb128: are you sure to run the new nautilus in that session? [10:51] Trevinho: if I call RequestReboot() from g-session through d-feet or gdbus, it works fine [10:51] Trevinho: but not if I call it from the indicator, which makes me wonder what happens in addition [10:51] Trevinho, yes [10:52] pitti: the indicator is not calling RequestReboot now, but Reboot since this implies a call to the org.gnome.shell EndSessionDialog that we implement [10:52] Trevinho, as said, if I do "nautilus /media/usb_stick" from a command line it doesn't open a duplicated view [10:52] Trevinho: same difference, that also works [10:52] Trevinho: I wondered if something killed the session before it calls [Request]Reboot()? [10:52] pitti: I'm wondering if logind is talking with gnome-session and if this changes something on the behavior we have [10:53] Trevinho: I'll log session and system d-bus requests, try logout from the indicator, and see what happens [10:54] Trevinho: btw, if I install consolekit, the indicator restart does work again [10:54] brb [10:55] pitti: mh, ok [10:56] Trevinho, hum, if I open the trash and close the view on it I get some warnings on stdout/err [10:56] (nautilus:26549): GLib-GObject-WARNING **: instance with invalid (NULL) class pointer [10:57] seb128: Mh, I check them [10:57] Trevinho, that was already happening before you patch it seems (just downgraded and still got that) [10:57] seb128: ah, ok... [10:57] well, in fact I get [10:57] (nautilus:27254): GLib-GObject-WARNING **: instance of invalid non-instantiatable type `' [10:58] with the old version [10:58] I check [10:58] Trevinho, what is running the commands when you click on an icon from the launcher? [10:58] I tried to strace compiz to see what nautilus command is called [10:58] but that doesn't work [10:59] doh [10:59] seb128: g_app_info_launch_default_for_uri with the mount location [10:59] I used the wrong pid [11:00] Trevinho, [pid 27938] execve("/usr/bin/nautilus", ["nautilus", "--new-window", "/media/seb128/e5654efa-864c-4efc"...], [/* 48 vars */]) = 0 [11:00] Trevinho: meh, this is quite ridiculous -- I tried four times after a clean boot, and four times it failed and threw me back to lightdm [11:00] Trevinho, does it use the .desktop to get the command? [11:00] Trevinho: then I tried four times with dbus-monitor attached, and it always worked [11:01] Trevinho: heisenbugs at its best! [11:01] seb128: [11:01] seb128: yaeh, that's the problem [11:02] Trevinho, wonder why it doesn't do it for you then? [11:02] hm, and now my unity/compiz settings are ruined [11:02] pitti: oh [11:02] seb128: because I've not installed the new .desktop file! :/ [11:02] pitti: and it's not Tuesday! :) [11:02] Trevinho, I see, good, at least it makes sense now [11:03] seb128: so... how can we make sure that the --new-window is not used when opening with a parameter, just using the .desktop file? [11:03] didrocks: indeed, so that's not it :) [11:03] seb128: otherwise the other way we have is adding some special code to untiy to pass that "--new-window" parameter only on middle click... but I would have avoided it [11:03] meh, I want my keybindings back [11:04] smells like bug 1042041 again [11:04] Launchpad bug 1042041 in compiz (Ubuntu) "1:0.9.8+bzr3319-0ubuntu1 regression: keeps setting gsettings keys to wrong values" [High,Fix released] https://launchpad.net/bugs/1042041 [11:04] Trevinho, let me think [11:04] but that'll be history with the unity rewrite anyway [11:05] Trevinho, we could try to make /usr/share/applications/nautilus-folder-handler.desktop be used [11:05] seb128: yeah [11:08] seb128: however if we really can't do it, I'd just remove the --new-window as default from the .desktop file, keeping it only on the quicklist [11:09] Trevinho, it would mean that running nautilus from the dash wouldn't start a new view if you already have one open though [11:09] which is a bit weird [11:10] mh [11:13] Trevinho, I wonder if just changing /usr/share/applications/defaults.list inode/directory=nautilus.desktop to nautilus-folder-handler.desktop would fix it? [11:14] seb128: it should [11:14] seb128: can you give a try? [11:15] Trevinho, that seems to work [11:15] Trevinho, let me do that change then [11:15] seb128: wow [11:15] Trevinho, we still have focus issues though :-( [11:16] seb128: how it's made it? [11:16] seb128: are you on unity dev? [11:16] no, raring [11:16] seb128: ok so wait for the update [11:16] ok [11:16] seb128: it's arleady in trunk [11:16] Trevinho, "how it's made it?", you mean? the defaults.list? [11:16] seb128: yes [11:16] Trevinho, it's a static list we ship in desktop-file-utils package [11:17] seb128: ah. ok [11:17] Trevinho, thanks for work, and glad we found an easy solution to the .desktop issue ;-) [11:17] seb128: np, thank you for looking into it :) [11:19] seb128: wouldn't be better to have a NN_git_* naming for the git patches as well? [11:20] seb128: since in this way applying them in order doesn't make a workable nautilus at all the stages [11:20] seb128: I mean, working on bzr+ssh://bazaar.launchpad.net/+branch/ubuntu/nautilus/ you can't push/pop the patches and always having a compilable nautilus [11:22] didrocks: one thing I always forget, if I open a bug for upstream unity, will the merger automatically include the distro bug when merging/releasing or not? [11:22] didrocks: I mean, will it show in the changelog? === chaoticuk_ is now known as chaoticuk [11:25] Trevinho, the series define the order [11:25] Trevinho: if you mention in the commit "bug #12345" of anything similar to it, or if you bzr commit --fixes lp:12345 or if you link to the MP [11:26] Launchpad bug 12345 in isdnutils (Ubuntu) "isdn does not work, fritz avm (pnp?)" [Medium,Fix released] https://launchpad.net/bugs/12345 [11:26] Trevinho, so no need of the nn_ [11:26] or* [11:26] Trevinho: those are the 3 ways to have the change mentionned in the changelog [11:26] seb128: yeah, but the name helps [11:26] Trevinho: the linkage to the MP should be done BEFORE it's merged though [11:27] didrocks: yaeh, but I mean... I open a bug #1234567 on lp:unity (only), will bhe the lp:ubuntu/+source/unity bug added automatically? [11:27] Trevinho, good point, will keep the numbering next time (I just like to have the git_ info since it makes easier to filter out things that are upstream or not) [11:27] Trevinho: if the bug is linked one way or another to a commit in the branch, yeah, downstream is opened automatically [11:27] seb128: yes, it makes sense [11:28] didrocks: ok, fine thanks [11:28] yw! [11:28] didrocks: as I don't always remember to do both (but I try) [11:28] Trevinho: no worry, I'm doing it for you automatically in that case :) [11:28] * Trevinho loves didrocks' bots [11:28] :-) === MacSlow is now known as MacSlow|lunch [12:24] non! il neige à nouveau [12:25] quand va-t-il terminer ? [12:29] pitti: au moins, nous avons la chance de travailler à domicile! ;) === alecu_ is now known as alecu [12:38] pitti: toujours un grand soleil ici et ciel bleu [12:38] pitti: mais beaucoup de vent [12:39] Trevinho: vraiment? tu parles français aussi? [12:39] ***headlines*** [12:39] :-) [12:40] didrocks: oui.. Je l'ai étudié à l'école, mais je ne le rappelle plus très bien... [12:41] Trevinho: assez bien tout de même [12:41] Trevinho: if you choose me as a performance peer reviewer, you just won an extra bonus point :) [12:41] * Trevinho adds him :D [12:42] ;) [12:43] Trevinho: you were not kidding! :) [12:45] didrocks: ahah, non... Je ne perdre pas les occasions :) [12:45] ;) [12:45] pitti, ici aussi il neige :-( [12:45] didrocks, t'en fais pas, je t'envoie la neige, devrait être là demain :p [12:45] seb128: t'as déjà essayé hier pendant que je courrais! [12:45] c'était vil et mesquin! [12:46] ;-) [12:46] thanks god, Lyon is a good city to live in :) http://france.meteofrance.com/france/meteo?PREVISIONS_PORTLET.path=previsionsville/691230 [12:46] * Trevinho sais que ubuntu-desktop sera en francais... [12:47] Trevinho: how many phoronix posts once we've done that? :-) [12:47] OMG ... all that phablet porting makes people speak in tongues ! [12:47] heh [12:48] ogra_: then, your tablet will start to be on strike :) [12:48] haha [12:48] "but but, I want you to start" [12:48] "no" [12:48] "non" you mean :) [12:48] and the phone will tell that they finished working [12:48] ogra_: right non :) [12:48] "c'est la fin de journée" === vrruiz_ is now known as rvr_ === MacSlow|lunch is now known as MacSlow === m_conley_away is now known as m_conley [14:02] kenvandine: I would like your advice. https://bugs.launchpad.net/ubuntu/+source/chromium-browser/+bug/1153137 [14:02] Launchpad bug 1153137 in chromium-browser "Please remove recommends on webaccounts-chromium-extension and unity-chromium-extension" [High,Triaged] [14:04] also lp bug#1154218 [14:04] * qengho slaps ubot2 === vrruiz_ is now known as rvr_ [15:19] libEGL warning: GLX/DRI2 is not supported --> hm, Xephyr in raring is still broken :( [15:25] xclaesse, is Xephyr supposed to support glx? [15:26] seb128, on quantal I had to backport precise's package [15:26] seems on raring it's still not fixed [15:26] precise is older [15:26] there is a bug somewhere for that [15:26] backport is the other way around usually :p [15:26] seb128, older but it works :) [15:26] what package did you have to backport? only the xephyr one? [15:27] yep [15:27] IIRC the bug was fixed upstream for months [15:28] tjaalton, mlankhorst: ^ know about that? [15:28] actually raring has the same version than quantal [15:28] so it's not surprising [15:30] nobody has bisected it though [15:30] seb128, voilà, just dpkg -i the .deb from precise and it works now [15:30] i think [15:32] ohh, compiz now takes org.gnome.Shell dbus name [15:32] so starting gnome-shell in xephyr on the same session won't work [15:32] https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/1079096 [15:32] Launchpad bug 1079096 in xorg-server "Xephyr does not have GLX" [Undecided,Confirmed] [15:32] that was working on Quantal :/ [15:33] mlankhorst, yep that's the one I reported :) [15:33] mlankhorst, right, that's the bug xclaesse opened (that got ignored so far) [15:34] yeah but its a bug upstream, blergh I'll take al ook at it later, g2g [15:34] mlankhorst, thanks [15:37] mlankhorst, thanks :) [15:55] hey larsu, question for you in https://code.launchpad.net/~larsu/python-dbusmock/signal-method-calls/+merge/153376 [15:58] larsu: so this looks a bit redundant, but if it's useful for you I'm fine with merging it (except one detail, see question) [16:02] chrisccoulson: ping? [16:04] chrisccoulson: could you drop me a hint how firefox organizes its headers and -I paths? do they collect the header at some 'central' include dir, do they sprikle the headers all over the source tree or do they organize in some more local modules or somesuch ... [16:15] pitti: argh I forgot to remove the log, I used that for debugging [16:15] larsu: ok, I thought something like that [16:15] larsu: I'm interested in how a signal is easier than GetCalls() or GetMethodCalls()? [16:17] pitti: I have an API that batches dbus calls and sends them later in an idle. In the test, I have to poll on GetMethodCalls() and that seemed a bit inefficient [16:17] having a signal and spinning a mainloop for a couple of seconds makes that much easier [16:17] I'm not 100% sure yet, wanted to hear your opinion about it [16:18] larsu: oh, I'm ok with that, I was just curious [16:18] cool :) [16:19] pitti: want me to remove the log or are you removing it on the merge? [16:20] larsu: I can remove it during merge, no worries [16:20] pitti: thanks! [16:23] what's the Quantal way to turn off auto-raise? this doesn't appear to work any more: gsettings set org.gnome.desktop.wm.preferences auto-raise false [16:24] sarnold, hey, did your systemd review include systemd-services (hostnamed, timedated)? [16:24] seb128: no [16:24] sarnold, hum, k, any reason it didn't? ;-) [16:25] sarnold, the MIR was first opened for those [16:25] seb128: jdstrand asked me to hit logind and skim udevd [16:25] ok [16:25] jdstrand, do you want the services security team reviewed as well? [16:25] seb128: only because I didn't notice those [16:26] seb128: basically, since the codebase is big, we were trying to focus on what you needed [16:27] seb128: I'm guessing those run as root and process input from users in some way? [16:27] jdstrand, yes [16:27] https://bugs.launchpad.net/ubuntu/+source/ubuntu-system-service/+bug/1153567 [16:27] Launchpad bug 1153567 in ubuntu-system-service "[FFE] Use systemd-services rather than ubuntu-system-service systemdcompatibility code" [Low,Triaged] [16:27] sarnold: sorry, would you mind taking a look at those too? ^ [16:27] jdstrand: sure [16:27] sarnold: I appreciate it [16:27] jdstrand, sarnold: thanks, sorry for the confusion [16:28] that MIR scope expended a bit [16:28] well, "- in a first step we want systemd-services promoted to replace ubuntu-system-services" was the first item in the rational [16:28] seb128: np :) [16:28] but the discussion got hijacked by the logind one ;-) [16:29] mterry, if I understand correctly your comment on the MIR you will ack it once the unit tests are running? [16:31] hey desktoppers, I've lost my ability to change workspaces in Raring with Ctrl+Alt+arrows, not sure if an update caused it. Any ideas on what I could look for to try and find out what's going on? [16:33] seb128, yeah, or they are shown to be not-usable at all for Ubuntu for some reason. Just wanted the situation investigated. pitti says he found out why they aren't working [16:33] they depend on /etc/machine-id, I'm filing a bug ATM === chaoticuk_ is now known as chaoticuk [16:34] https://bugs.freedesktop.org/show_bug.cgi?id=62344 [16:34] Freedesktop bug 62344 in general "tests depend on /etc/machine-id" [Minor,New] [16:48] sarnold: erf, there is also this one that makes sense since you are looking at the other bits: 1153633 [16:48] bug 1153633 [16:48] Launchpad bug 1153633 in systemd-shim (Ubuntu) "[MIR] systemd-shim" [Undecided,New] https://launchpad.net/bugs/1153633 [16:49] jdstrand: "erf" indeed :) hehe [16:49] sarnold: sorry man, this MIR is spinning out a bit [16:49] hopefully that's 'it' [16:50] hehe [16:50] sarnold: hi [16:50] sarnold: but think of it this way, at least you aren't do bi-weekly openjdk updates /o\ [16:50] jdstrand: hehe, yes, that thought crossed my mind :D [16:50] :) [16:51] hey desrt :) [16:51] i'm around for any questions :) [16:51] and i also appreciate feedback on things that you think i could do to improve security [18:04] I'm having this problem with my overlay scrollbars sometimes not showing the clickable thumbbar. Is that just due to me being on gtk 3.7? [18:04] (i.e. are other people seeing that?) [18:04] * didrocks waves good evening [18:04] mterry: on the very edge of the screen? [18:04] if your app is maximized? [18:05] didrocks, no, this could be anywhere, not the "pixels on right click behind" issue [18:05] mterry: ok, I tried to link to that one, indeed :) [18:05] didrocks, (is the edge-of-the-screen issue only in gtk 3.7?) [18:05] That's been annoying too [18:05] mterry: the fix is merged [18:05] Cool [18:05] mterry: just waiting for a release with tests passing ;) [18:05] didrocks, you waved good evening, get out of here [18:05] mterry: or you rerunning at the failure was this morning [18:06] :) [18:06] as* [18:06] due to UTAH [18:06] so either publish tomorrow, either rerun I guess [18:06] and yes, I'm already out [18:06] :-) [18:06] * didrocks byes for real now === alan_g is now known as alan_g|life [18:09] kenvandine: what did you need autopilot-qt for? do you need to (build-) depend on it for some package? [18:10] (i'm writing the FFe, and making sure I don't say things that aren't true) [18:10] mterry, I didn't notice issues with the scrollbars (using 3.7 as well), but I'm not a bit scrollbar user (I tend to scroll with the scrolling wheel) [18:10] or with keyboard page up/down [18:10] seb128, hrm. try using it a bit more and see if you hit anything [18:12] mterry, yeah, I can't make it appear at all on my gedit win atm [18:13] seb128, hrm, good that it's not just me then [18:14] seems to wfm on 3.6 [18:15] mterry, yeah, gtk 3.7 issue it seems [18:15] mterry, ld_library_path loading 3.6 and it works [18:15] it does it every time on that .c I had open in gedit [18:15] it shows the thumb on first try [18:15] then it stops working [18:16] seb128, yes. I've noticed that it always seems to work at least once. It seems per-app not per-window [18:16] mterry, I'm glad I didn't try to force the new GTK in :p [18:16] Laney, good, at least raring is safe [18:17] seb128, probably should have asked about this earlier. :) I've been seeing this for a while [18:17] mterry, going to be fun to track down [18:17] I doubt anyone in the GTK world is interested to look at overlay scrollbars [18:17] and I doubt anyone in here is interested by looking at overlay scrollbars bugs [18:18] mlankhorst, btw, for info, xephyr/glx issue debugged, patch attached to https://bugs.launchpad.net/xorg-server/+bug/1079096/comments/6 [18:18] Launchpad bug 1079096 in xorg-server "Xephyr does not have GLX" [Low,In progress] [18:19] mlankhorst, not that I've no clue what I'm doing there, I'm not an xorg hacker, I just copied a commit that fixed a similar issue for vfb :p [18:26] seb128, who are our overlay experts? [18:27] mterry, I saw you just raised your hand? :p [18:27] seb128, oh man [18:27] seb128, I'll look into it, no problem, but sounds like this is post raring worry? [18:27] mterry, joke aside Cimi did most of the work but I don't think he's interested by maintaining the old stuff he worked on... [18:27] you can still ask him [18:27] but I guess he will say it's a gtk issue [18:28] seb128, we aren't trying to squeeze 3.8 into raring anymore, right? [18:28] the other ones who looked at some issues there/helped him are kenvandine and desrt [18:28] mterry, I don't plan to, there are still some known bug (like the screenshot one), and there is strong reason to do it [18:28] +no [18:28] yup :) [18:29] seb128, OK. I've added it on the bottom of my TODO list, let's see [18:29] mterry, thanks [18:29] maybe desrt or kenvandine can help having a look [18:29] but I guess they are both not so interested and will try to escape it ;-) [18:36] cyphermox, i want it for testing qml-friends and gwibber [18:41] dpm, pitti: we didn't get any langpack upload in raring or am I overlooking something? [18:41] is there any planned soon? === dpm is now known as dpm-afk === rickspencer3_ is now known as rickspencer3 [19:25] Laney, I'd like to get early access to the glib package that you uploaded yesterday. I tried looking for it in the pool, but apparently it's not yet there, is it accessible somewhere? [19:34] marga, what "pool" did you use? [19:34] marga, it's in the queue: https://launchpad.net/ubuntu/precise/+queue?queue_state=1 [19:34] things don't build before they are accepted though [19:35] seb128, http://us.archive.ubuntu.com/ubuntu/pool/main/g/glib2.0/ [19:35] so you need to wait for the SRU team to ack it or build it locally [19:35] right, I was thinking of building it locally [19:35] marga, you have the files links in the queue ui, you can just download those [19:35] great, thanks. [19:36] queue is the gateway before the archive [19:36] so it's normal that things don't land there before being reviewed === _thumper_ is now known as thumper === m_conley is now known as m_conley_away [22:02] RAOF, hi [22:02] tkamppeter: Hi. Ah, yes, I need to upload that. [22:03] RAOF, beta1 is over and GS is waiting for lcms2 ... === jbicha_ is now known as jbicha [22:26] tkamppeter: Actually, how did you generate the patch on the bug? [22:27] tkamppeter: Because it really seems to have a lot of unrelated changes. [22:42] RAOF, I have taken the lcms2-2.4 version which ships with GS 9.07 and diffed it with the lcms2-2.4 version of the .orig.tar.gz of the Ubuntu package. [22:43] RAOF, with this patch all the sample PDFs rendered without crash, and also nothing else crashed on the test system. [22:45] tkamppeter: Ah, so it's not in any way intended to be a minimal fix. Ok. [22:46] RAOF, it is intended to get the changes made by the GS developers into lcms2, as the upstream patch has failed, probably missing something.