=== AndChat690900 is now known as chucrute === rickspencer3_ is now known as rickspencer3 [03:00] darkxst: Is there any particular reason why the Ubuntu GNOME remix images load straight into the language/install boot menu, and not pause a little while like the Ubuntu images do? [03:00] darkxst: I would ask Jeremy, but he is not online atm. [03:01] TheMuso, not sure, we don't actually touch any of the boot/installer stuff (apart from the ubiquity slideshow) [03:02] so it should be the same as Ubuntu really [03:07] darkxst: Ok, well its not. If you were to boot a standard Ubuntu image, and then boot an Ubuntu GNOME image, you would notice that the Ubuntu iamge doesn't automatically bring up the language choice and the boot menu. There is a slight pause before the Ubuntu image boots directly into the try/install ubuntu ubiquity screen. If you were to press a key during this pause on the Ubuntu image, you would get what you get on the Ubuntu GNOME remix ima [03:08] darkxst: I'll talk to Jeremy about it when I see him online next. [03:10] TheMuso, right, I see what you mean [03:11] darkxst: The only reason why I mention it, is because it has accessibility implications WRT the easse of installing with Orca et al. [03:20] TheMuso, I will look into a bit later, but I guess its too late now to fix for the release [03:29] darkxst: Yeah unfortunately. [03:52] Good morning [03:57] pitti, good morning [04:02] pitti, any idea how I can get debug logs from g-s-d (on login)? its either not starting, or crashing [04:03] starting it from a shell doesn't crash? [04:03] normally you can start an xterm session, start gnome-session in one, and gnome-settings-daemon in another [04:03] or killall gnome-settings-daemon; gnome-settings-daemon in a running session [04:03] pitti, starting in a shell is fine [04:04] and nothing in .xsession-errors either, I presume? [04:04] normally output goes there [04:05] no I don't see any output in there [04:05] if you want debugging, you need to run everything under G_MESSAGES_DEBUG=all [04:05] you could define that in /etc/environment temporarily, or in your ~/.profile [04:06] but at at least a warning or two usually appears in .xsession-errors [04:06] (gnome-settings-daemon:2280): color-plugin-WARNING **: There is no colord server available [04:06] I have that one at least [04:06] darkxst: a deeper hack: replace /usr/bin/gnome-settings-daemon with a shell script that does something like this: [04:06] #!/bin/sh [04:07] G_MESSAGES_DEBUG=all /usr/bin/gnome-settings-daemon.real "$@" > /tmp/gsd.log 2>&1 [04:07] (and rename the binary accordingly) [04:19] pitti, oh its trying to connect to the shell dbus [04:19] and that isn't running yet? [04:19] I am running it in unity right now [04:20] but it sounds like it should handle that gracefully [04:20] and unity spoofs the dbus name "org.gnome.Shell"? [04:21] pitti, (gnome-settings-daemon.real:1785): media-keys-plugin-WARNING **: 0: Cannot invoke method; proxy is for a well-known name without an owner and proxy was constructed with the G_DBUS_PROXY_FLAGS_DO_NOT_AUTO_START flag [04:21] g_dbus_connection_real_closed: Remote peer vanished with error: Error sending message: Broken pipe (g-io-error-quark, 44). Exiting. [04:21] ek, indeed [04:33] pitti, something is triggering a SIGTERM, causing g-s-d to shutdown [04:34] hm, could that be upstart? did you enable upstart sessions? [04:35] pitti, I am using normal upstart configuration [04:35] well, that hasn't been enabled by default yet [04:35] can you try without and see if that makes any difference? [04:36] how do I disable that? [04:36] darkxst: I hear that ~/.cache/upstart/*.log should have logs then, do they show anythign interesting? [04:36] how did you enable it? [04:36] * pitti looks for jodh's blog post [04:36] pitti, I have not touched anything upstart [04:36] http://ifdeflinux.blogspot.de/2013/04/upstart-user-sessions-in-ubuntu-raring.html [04:37] ah, so you are not using session upstart then [04:37] darkxst: so perhaps try the debugging wrappers around gnome-session and gnome-settings-daemon? perhaps session is killing it for some reason [04:43] ok will try [04:53] pitti, nope, can't see anything interesting in the gnome-session logs [05:46] ricotz, hi [06:36] hello [06:37] 13.04 is still not out ???? [06:43] can I disable sudo mail somehow? [06:47] Y [06:47] mlankhorst: a few guesses.. set mail_no_user off or set mailerpath to /bin/true or /bin/false [06:53] hm found it, I think === tkamppeter_ is now known as tkamppeter [06:53] Defaults !mail_badpass [06:53] same for the other mail flags [07:03] good morning [07:08] salut jibel! ça va? pas trop le dos cassé avec le déménagement? [07:11] Salut didrocks! Etonnamment j'ai le dos entier, merci :) [07:11] bien content que ce soit terminé [07:12] j'imagine :) [07:12] ça s'est bien passé? [07:18] didrocks, oui parfait. Il reste à tout déballer si je veux pouvoir m'habiller ;) [07:19] ahah, /me reminds himself to not do any hangout with jibel today :p [07:29] hey desktopers [07:31] happy release day seb128! [07:31] oh, right [07:31] didrocks, happy raring day ;-) [07:34] I feel very raring today! [07:34] bonjour seb128, ça va ! [07:35] hey BigWhale [07:35] salut didrocks [07:35] hey pitti! [07:35] pitti, salut, oui, il fait beau ici, et toi ? [07:35] seb128: ici aussi, il y a soleil ! [07:36] le temps parfait pour le release :) [07:36] oui ;-) [07:36] hey seb. :) [07:36] and the rest of the gang :) [07:36] didrocks: veux-tu courier à nouveau la semaine prochaine ? [07:36] * pitti waves to BigWhale [07:36] pitti: avec joie! J'espère que tu traineras seb128 pour qu'il vienne avec nous :) [07:37] pitti: je trainerai larsu aussi ;) [07:37] didrocks: (ou est-il "à la semaine../au semaine" ?) [07:37] no way I go running in Oakland [07:37] I wish I had google translate plugin for xchat ... :> [07:37] I don't like running in city, less so in not so nice ones [07:37] pitti: c'était très bien: "veux-tu aller courrir à nouveau la semaine prochaine" [07:37] seb128: there is a park [07:38] and crazy people yelling at you in the street? ;-) [07:38] well, that's a small detail :) [07:38] didrocks: ah, "aller courrier" est mieux ? [07:38] pitti: "aller courrir", oui :) [07:38] courrier == mail [07:39] seb128: non, c'était bon -- beaucoup de gens le font [07:39] pitti, on verra, j'essayerai peut-être [07:40] viens avec nous! [07:40] Didier m'a dit que la gym a l'hotel n'est pas bien en plus [07:40] seb128: il va être facile [07:41] pitti, tu arrives à suivre Didier ? il s'entraîne tous les jours de l'année [07:41] je ne vais pas courrier beaucoup [07:42] seb128: no, we run together; he's nice with me :) [07:42] ;-) [07:42] let's try [07:42] \o/ [07:42] "il est gentil avec moi" ? [07:42] oui [07:42] vous arrivez tous les 2 le dimanche ? [07:43] dimanche pour moi [07:43] j'ai arriver à samedi soir [07:43] pitti, ah, moi aussi [07:44] seb128: ah, alors nous pouvons aller à San Francisco à dimanche ? [07:44] oui [07:44] \o/ [07:44] * seb128 donne l'accolade à pitti [07:44] * pitti tu donne une accolade en retour [07:44] "te" [07:46] * pitti goes offline for a bit for some NM tests [07:47] good luck ;-) [08:07] hey [08:08] Laney, good morning, happy raring day! [08:08] \o/ [08:09] * Laney starts spamming "is it out yet?" every hour [08:09] Laney, 'spamming' would be every minute! [08:10] FINE, YOU ASKED FOR IT! [08:10] Laney, ok, you do that, and I will do the "is the new serie open yet?" ;-) [08:10] TheMuso: good spot. Probably grub/isolinux timeout is not in, or very short with a black screen =) [08:11] seb128: We probably need to know what it's called first :P [08:12] morning [08:13] o hai [08:13] mlankhorst: cycled 35 miles last night, got lost and had to get a train home [08:14] hehehe [08:14] I have a pick up service for that! :D [08:14] daddy? [08:14] Laney, don't you have a gps! [08:14] only for emergencies! [08:15] I did know where I was going but the getting lost made me too late [08:15] hehehe [08:15] Laney: yeah he has a bus, I can just put the bike in the back and leave it standing there [08:15] heh [08:16] sometimes I do it the other way around too when he gets a flat tire [08:17] mlankhorst, flat tire? what is that, I havent had one ever since running tubeless ;) [08:17] don't worry, I'm sure your bike will still have a failure mode [08:18] mlankhorst, sure, and if that happens I will be walking, since not many roads where we ride [08:19] yeah if it really came to it my cycle insurance includes pick up [08:19] psssh insurance for bikes [08:19] Laney, helicopter pick up? [08:20] lolz [08:20] * Laney does road cycling [08:20] mlankhorst: you need it if you live where i do :p [08:21] I just made my bike look unattractive and valueless instead.. [08:21] mlankhorst, that doesnt always work, but atleast my bike always looks clean, since its the same colour as the mud! [08:22] darkxst: plus it looks unique enough that nobody dares to steal it :-) [08:23] oh that wouldnt work here [08:23] the thieves here will steal it and try and sell it back to you! [08:24] so lets say you have a street view like http://goo.gl/maps/5DQsv [08:24] would you steal a bike that looks like any other or the one that makes you stand out a lot? [08:30] xnox, what is the release-?-incoming tagged used for? [08:30] xnox, I don't see it listed on https://wiki.ubuntu.com/Bugs/Tags [08:31] seb128: to get it on http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-r-incoming-bug-tasks.html [08:33] seb128: the intention is to end up on the rls-*-tracking report. as per Laney. But I sometimes fail to spell the tag correctly. [08:34] Laney, xnox: thanks, I had almost forgotten about that tracker [08:34] oh you actually did 'release-' [08:35] Laney: please correct the spelling to the right one =) i did ping bmurray to add those as official tags with auto-complete. [08:35] no idea how that stuff works [08:40] pitti, any idea how I can get gdb attached to gnome-settings-daemon at login? [08:41] what are you trying to get? [08:41] why do you need attached at login? [08:41] seb128, gnome-settings-daemon 3.8 dies on login when running unity [08:41] but runs perfectly after logged in [08:41] but apport doesn't give you a dump you can use gdb on? [08:41] seb128, nope, seems that g-s-d is shutting itself down [08:42] if it does, gdb will not help you [08:42] seb128, g_dbus_connection_real_closed: Remote peer vanished with error: Underlying GIOStream returned 0 bytes on an async read (g-io-error-quark, 0). Exiting. [08:42] seb128, there is a SIGTERM signal atleast when running from gdm [08:43] darkxst, do you get the issue if you run g-s-d in an empty session? [08:43] like [08:43] Xephyr :1 [08:43] DISPLAY=:1 gnome-settings-daemon [08:43] with dbus-launch in front if needed [08:43] it might be a "start before gnome-session" or something [08:45] darkxst: you could extend you debug wrapper to run xterm -e gdb ... g-s-d ? [08:45] I have never used Xephyr but it just gives me an empty window [08:45] darkxst: are you sure that g-s-d is shutting *itself* down? [08:45] darkxst: I thought you said something sent it a SIGTERM? [08:46] pitti, SIGTERM is only when launching from gdm [08:47] darkxst: right, if you start Xephyr you just get an "empty" X server, you have to run stuff in it [08:48] ok a few warnings but otherwise fine [08:49] ok, so it's not a "it needs other components to be ready when they are not" [08:55] apparently not, but something is causing it too shutdown [08:56] I can install the 3.6 ck built g-s-d and it survives login just fine (even though I am running logind) [09:04] darkxst: what version of gsd are you using? [09:04] I just dist-upgraded to gnome3 and gnome3-staging in a VM and I can still get unity [09:04] Laney, 3.7.91 (and 3.8.1) [09:05] you will get unity, but gsd is dead by the time you are logged in [09:07] so I added xterm to my debug wrapper, but it only shows up before the gdm session, [09:11] should the same gsd instance persist or does it get respawned on login? [09:11] each session has its own gsd [09:11] that is what I thought [09:12] oh it comes up when I logout ;( [09:12] WTH [09:12] darkxst: I guess you need similar debug logs for gnome-session to determine if and when it starts g-s-d [09:12] could it be that the gdm one doesn't exit and is running on the same display or something? [09:12] dont install logind backwards :) [09:14] seb128, maybe its just at the back of the stack and not visible [09:14] * ogra_ guesses what you see going down and starting up is the DM g-s-d [09:14] check which user owns the process [09:15] ogra_, do you run gdm? [09:15] *processes [09:15] nope [09:15] darkxst, does it happen if you start the session with startx? [09:22] seb128, right now startx fails to load the session [09:22] X is running though [09:23] have an .xinitrc with "exec gnome-session" or similar? [09:27] seb128, do now, all I get is keyring prompt [09:40] seb128, I can only start gnome session with startx not ubuntu session [09:40] weird [09:57] pitti, did we end up having a flag or something in the apport reports that tell us if the issue happened in a session which is not the current one? [09:57] pitti, I'm looking at https://errors.ubuntu.com/problem/8198ebba906a1181a84ed08ff814f54c7c737ea3 (udisks2 segfault) [09:57] pitti, on the corresponding launchpad bug you asked by then if the error was displayed on login [09:58] seb128: for programs that run in the session we get XsessionErrors if the program that crashed and apport are running in the same session [09:58] for daemons we won't, as something like udisks has no idea about user sessions [09:58] right [09:59] the comments suggest it's happening a login but also sometimes after trying to mount devices [09:59] I keep getting dupes of that, but I never could reproduce it (and I tried for an hour or so) [09:59] pitti, it's still ranked quite high on e.u.c for raring [09:59] it's n°25 of raring errors for the last 10 days === MacSlow is now known as MacSlow|lunch === e11bits_ is now known as e11bits === sil2100_ is now known as sil2100 === asac` is now known as asac [11:54] seb128, anyway I am basically over this for today, I think my patch fixes the media key issues under unity (i.e reverts the legacy key grabber for unity). see gsd package here https://launchpad.net/~darkxst/+archive/logind [11:55] seb128, probably would be better to move that all into a seperate module, in the long term if you plan to leave it in gsd [11:56] darkxst, great [11:56] we probably wont keep it in g-s-d longer than for the next lts [11:56] unity-next will probably handle keybindings like gnome-shell does for GNOME [11:57] yes that makes more sense, gnome-shell has a lot issues with special key grabs previously [11:57] s/has/had/ === alan_g is now known as alan_g|lunch === attente_ is now known as attente [12:37] seb128: hey [12:37] good morning [12:37] desrt, hey, how are you? [12:37] okay [12:37] seb128: i noticed you postponed some of my workitems [12:37] desrt, right, raring has been released [12:37] seb128: i was wondering if there is some search interface that lets me find all blueprints with my name on them for work items [12:38] because i had completely lost track of those items [12:38] desrt, https://launchpad.net/~desrt/+upcomingwork [12:38] ^ note that this won't show WIs for past milestones [12:38] perfect. [12:38] thanks :) [12:38] http://status.ubuntu.com/ubuntu-raring/u/desrt.html for that [12:38] right, pitti beat me to it [12:39] I have +upcomingwork as my first firefox tab, it's a great todo list [12:39] desrt, if you remember, I got you added to some team last cycle so you would have that status.ubuntu.com page :p === MacSlow|lunch is now known as MacSlow === alan_g|lunch is now known as alan_g [13:20] happy release day! [13:22] kenvandine, same to you! :) [13:25] kenvandine, good morning, happy raring day to you ! [13:26] i'm raring to go~ [13:26] :-D [13:29] hey kenvandine, happy release day! [13:34] kenvandine: you reverted a lot of changes from cyphermox it seems, did you refresh before editing the blueprint? [13:34] oh sigh... i didn't [13:34] silly website... [13:34] kenvandine: do you have the email to reconcile? :) [13:34] no [13:34] kenvandine: sorry, I didn't know you were editing it too :) [13:34] doesn't matter, I'll go back and re-do any changes [13:35] thanks cyphermox :) [13:35] cyphermox, sorry... :) [13:35] that's the components thingy blueprint right? [13:35] i wonder how many times i've done that to people [13:35] delivering-touch-apps? [13:35] i never considered needing to refresh the page :) [13:35] cyphermox, yes [13:37] amazing... after living with evolution crashing hourly for months... i used geary all week without a single crash [13:37] * kenvandine is happy [13:37] haha [13:37] can we fix evolution? how is it crashing? [13:37] didrocks: speaking of the the same blueprint, isn't the icons merging done? [13:38] cyphermox, i don't recall the bug #, but the crash i was getting matched a bug that was like 2 years old [13:38] mirrors are having a rough day :) [13:39] desrt, happy release day! [13:39] :-D [13:39] kenvandine: same to you :) [13:39] i'm happy launchpad seems to handle it better these days [13:39] i remember jaunty, i couldn't do anything on launchpad all day [13:40] cyphermox: it should be yeah [13:42] tkamppeter: I posted instructions on how to valgrind xserver on the bug [13:43] kenvandine: be nice to lP :) [13:44] czajkowski, i am... it works on release day now :) [13:44] it didn't back then [13:49] seb128, hey, is there anything that needs to be done to revert that commit to fix the libreoffice bug? [14:13] mlankhorst, I have seen the instruction. Do I nbeed to install all 30+ -dbg packages of X or only the ones concerning my hardware? [14:14] evdev-dbg, synaptics-dbg, libdrm*dbg and your video driver debug is usually enough [14:24] mlankhorst, all set now. I have installed all -dbg packages you asked for, installed the script and linked it, and finally restarted lightdm, which came up much slower now. [14:24] ;P [14:25] well if it crashes you should have a full valgrind log at least [14:25] mlankhorst, from now on I will attch the log file to the bug report whenever a crash happens. [14:26] mlankhorst, but there is one problem: When a crash happened X gets automatically restarted (I get the lightdm screen). Will this not delete the log? [14:26] it's appending, rather than overwriting === alan_g is now known as alan_g|tea [14:29] mlankhorst, thanks, and how do I proceed if X gets stuck but does not crash? "sudo stop lightdm", grab log and attach to bug, "sudo start lightdm"? Or is the log useless when there was no crash? [14:30] just send a killall Xorg :p [14:31] you can use a debugger with valgrind too, but it's a bit more effort to set up [14:31] anyway I guess you mean the touch being stuck? not much you can do about that [14:31] only documenting how you got it stuck would be helpful === alan_g|tea is now known as alan_g [14:50] mlankhorst, onboard stopped working since I am running under Valgrind. [14:50] mm, any useful warnings in valgrind¿ [14:56] mlankhorst, the Valgrind log so far is http://paste.ubuntu.com/5601259/ [14:56] sigh intel doesn't have valgrind enabled, nothing harmful then :/ [14:57] attente, hey [14:57] attente, do you agree the revert is right? [14:57] attente, just filing a merge request against lp:unity/7.0 (that's the raring serie), can you do it? [14:58] tkamppeter: only thing sort of useful is that activatekeyboardgrab/pointergrab, but it looks like libunwind info is missing there [14:59] seb128, the revert is correct [14:59] seb128, jasoncwarner, everyone ... [14:59] congrats on 13.04 [14:59] ok, will file [14:59] It's my favorite Ubuntu yet [15:00] rickspencer3, thanks ;-) [15:00] attente, ok, thanks [15:00] :o) [15:01] mlankhorst, do I need to install a dbg package of libunwind? [15:01] nah [15:02] it was supposed to give a perttier backtrace === Sweetsha1k is now known as Sweetshark [15:04] * Sweetshark thinks about getting his saucy salamander in position ... [15:06] mlankhorst, Now I have installed libunwind8-dbg and restarted lightdm, now onboard works. [15:07] tkamppeter: it doesn't matter, the backtrace suggested that you were hitting the same bug I was hitting.. [15:09] Updated Valgrind log attached to the bug report. [15:10] tkamppeter: valgrind logs are not going to be useful, except for crashes [15:35] mlankhorst, I have added some more comments to the bug report. [15:42] is there any outstanding issues with the nvidia-310 proprietary drivers that would cause me to use acpi=off at boot time? [15:43] and sorry, I forgot to include on 13.04 [16:06] tedg, Laney: you guys are familiar with libproxy? do you know why it spawn those shell processes to run pxgsettings and why those processes keep running? [16:07] seb128, No, I'm not. I was just noticing that thread. [16:08] seb128: No, seems weird. I've got quite a few too [16:10] We should look into why they don't get closed [16:14] mmm, yeah, it appers to be some constantly running thing to notify clients of changes [16:18] seb128: we could probably rewrite this to be a dbus service if necessary? [16:20] seems pxgsettings is currently c/glib so it might not be too hard [16:20] want to WI me for it? [16:24] hum, got disconnected, I was saying [16:24] Laney, tedg: ok, I wanted to check if you knew about it before looking, let me have a look [16:24] seb128: Laney volunteered ;) [16:24] did he? [16:24] basically [16:24] 16:20 < Laney> seems pxgsettings is currently c/glib so it might not be too hard [16:24] Laney, thanks! ;-) [16:24] 16:20 < Laney> want to WI me for it? [16:25] Laney, do you think you have the time this week? [16:25] *this* week? [16:25] Laney, if not I might have a look to it in the plane saturday [16:25] oh, to investigate? [16:25] yes [16:25] well I kind of understand what it does [16:25] ok [16:25] I'm letting it for you [16:25] if you just run the pxgsettings thing you have running on your system you can see [16:25] it watches for changes and outputs them onto stdout [16:26] and then libproxy reads those and presumably relays it to its clients [16:26] it just seemed a good offline/flight hack, but I've other stuff to keep me busy so no worry ;-) [16:26] ah [16:26] that seems hackish [16:26] Seems like you could replace that with a dbus service [16:26] yeah, that would be nicer [16:27] library that spawns shell that calls C binary that opens other file which can be accessed over DBus, nah, not hackish at all ;) [16:28] hey, maybe they've got a good reason ... [16:28] * Laney runs [16:28] maybe ;) [16:28] I start understanding why Josselin hates libproxy :p [16:29] IIRC glib-networking even falls back to libproxy these days if it needs to [16:30] so maybe most gnome stuff doesn't need to use libproxy explicitly [16:31] ah [16:31] seb128: see glib-networking/proxy/gnome/gproxyresolvergnome.c:307 [16:31] static gboolean [16:31] g_proxy_resolver_gnome_is_supported (GProxyResolver *object) [16:31] { return !g_strcmp0 (g_getenv ("DESKTOP_SESSION"), "gnome"); [16:31] } [16:32] Laney, we should archive grep for "DESKTOP_SESSION" [16:32] Laney, [16:32] libproxy/modules/config_gnome3.cpp: && string(getenv("DESKTOP_SESSION")) == "gnome")); [16:33] similar issue [16:33] well, that one checks for GNOME_DESKTOP_SESSION_ID in preference which means we're ok [16:33] but AFAICT that glib-networking check is why it's falling back to libproxy and therefore pxgsettings [16:33] * Laney test builds one with it fixed [16:33] oh [16:34] what would it use if that was not libproxy? [16:34] reads it directly from gsettings i think [16:44] Laney, you are right, that seems to fix it ;-) [16:44] great [16:44] Laney, do you know why glib-networking build-conflicts on itself? (that's annoying) [16:45] ah [16:45] no idea, maybe the testsuite? [16:45] yes [16:45] * Add build-conflict about glib-networking itself (fails the test [16:45] suite). [16:45] think we should SRU this? [16:45] Laney, do you want me to open a bug? [16:45] Laney, yes please [16:45] yes [16:46] seems like it was introduced in 2.35.3 which explains why it's new [16:46] * proxy/gnome: This is now only used in GNOME login sessions (as, [16:46] essentially, a more efficient version of the libproxy GNOME [16:46] backend); in non-GNOME sessions, gio will now fall back to the [16:46] libproxy plugin, [16:49] seb128: I still see them coming from signon{d,ui} though - do you? [16:49] Laney, I didn't restart those services, but they might use libproxy directly? [16:49] they do [16:54] Laney, https://bugs.launchpad.net/ubuntu/+source/glib-networking/+bug/1172792 [16:54] Launchpad bug 1172792 in glib-networking (Ubuntu) "should use the direct proxy resolver in unity sessions" [Undecided,New] [16:54] Laney, I've added SRU infos, I will let you deal with the patch/upload/upstreaming? [16:55] not sure why they want to limit gsettings use to GNOME [16:56]