[00:02] seb128: whoa, never knew about bzr bd-do, that's awesome [00:08] Me either, sounds very useful. [00:30] robert_ancell: i wonder if you know about this error i'm getting [00:31] disregard. i'm dumb. [00:31] was entirely related to the bad wifi we had :) [00:32] desrt, curious [00:50] robert_ancell, do I need anything in /usr/local ? i think it's causing program issues [00:51] bcurtiswx, ? [00:52] robert_ancell, i've built from source, i had it install in /usr/local .. i see other things and don't remember if i need anything from there (or can i delete it) [00:52] bcurtiswx, built what? [00:53] im keeping in general terms, i've built packages from source (telepathy-gabble for example) [00:56] bcurtiswx, I just crashed before - did you send me something before "robert_ancell, do I need anything in /usr/local ? i think it's causing program issues" [00:56] robert_ancell, i've built from source, i had it install in /usr/local .. i see other things and don't remember if i need anything from there (or can i delete it) [00:57] im keeping in general terms, i've built packages from source (telepathy-gabble for example) [00:58] bcurtiswx, oh, do you mean, "If I delete /usr/local will that kill anything required by Ubuntu?" [00:58] correct [00:59] robert_ancell, correct [00:59] You should be able to. I notice I've got some directories in /usr/local, I'm guessing they've come from me manually building something [01:02] robert_ancell, thx :) [01:07] Hm. I wonder if it'd be possible to shorten the skew between packages on archive.ubuntu.com and their corresponding -dbgsym packages on ddebs. [01:16] Oh, well. At least the compiz crash is easily reproducible :/. Local build, ho! [01:31] RAOF: You'd think they would be published at the same time as the other packages, or within an hour or so. [01:31] You might think that, yes. [01:32] That might be the case, but I apparently almost always want to install dbgsym packages within an hour of publication on a.u.c :) [01:38] heh [01:49] Man, no-one's reported this bug yet? Win! === asac_ is now known as asac [02:35] RAOF: heh === smspillaz|zzz is now known as smspillaz [03:26] Can anyone think of what might be creating a "Desktop" folder in the $HOME of a liveCD user of ubuntu? XDG's defaults are set to not create that folder, and nautilus's gconf is set to not use the desktop folder. And curiously, when running `adduser`, the folder is _NOT_ created. It is only on the liveCD desktop user. Any ideas? :/ [03:27] I'm trying to fix bitesize bugs from unity, but I'm in doubt on how to replace the current unity with my compiled one. Any help? [03:27] aroman: the Desktop folder has never been created by adduser. it's usually something that's part of the login process, but i can't remember what component [03:28] broder: Well, adduser runs skel,and iirc ubiquity has some custom adduser stuff going on [03:28] IanLiu: You can either build a package with your patch applied, or install your patched unity locally. [03:28] I mentioned that adduser worked as expected to confirm that it was a livecd only thing [03:29] IanLiu: I'd suggest doing the former; if you install it locally, it will override the system-wide one and so you won't get packaged updates. I had that happen for a while, and was wondering where all the new features other people were talking about were :) [03:30] RAOF: so, say I install it on /usr/local, I can replace the running unity with "/usr/local/unity --replace" ? [03:31] IanLiu: No; unity is a compiz plugin, so you need to put it somewhere compiz will look at. [03:32] IanLiu: That's either /usr/lib/compiz/ or ~/.compiz-1/(plugins?). The local-install target of unity's buildsystem will do the latter. [03:32] RAOF: I see. I will follow your hint and build the package ;-) [03:34] RAOF: so, after installing my patched package, will I need to restart X so my new unity is loaded? [03:34] You could do that. You could also just run “compiz --replace”, or “unity” (which essentially just runs compiz --replace for you) [03:35] RAOF: ahh, nice! [03:36] RAOF: last question for today ;-) How can I see my debugging messages? I'm writting some g_warning in unity source so I can understand what is happening, but I'm not sure how I will be able to read them [03:36] If you run “compiz --replace” (again, or unity) in the terminal, that's where the output will go, too :) [03:37] RAOF: Perfect. Thanks for your help! === Amaranth__ is now known as Amaranth [07:25] good morning all [07:42] Good morning [08:06] good morning everyone [08:07] hey chrisccoulson [08:12] hi pitti, how are you? [08:13] I'm fine, thanks! [08:13] chrisccoulson: got woken up early again? [08:13] pitti - yeah ;) [08:13] i have a really sore neck this morning [09:15] hey desktopers [09:18] seb128, heya [09:18] hey bryceh, how are you? [09:18] hey seb128, bonjour [09:18] hey pitti [09:19] seb128, doing good, you? [09:20] Aloha pitti, bryceh, seb128! [09:20] just finished reading latest dead horse flogging on planet [09:20] hey RAOF, good evening! [09:20] oh, new xorg fun landing [09:21] You're well, I trust? [09:21] bryceh, I'm fine, playing with the new unity ;-) [09:21] hey RAOF [09:21] Pay no mind to the new xorg stuff :) [09:21] I didn't notice much change, out of having recently used applications listed [09:21] seb128: Does the dash work for you? [09:21] yes [09:22] well I didn't use it, but it displays it's icons [09:22] waouh, new alt-f2 rocks [09:22] Hm. Maybe it's a local problem then; for me it segfaults compiz reliably. [09:22] it displays only one line now! [09:22] not the dash screen [09:22] the menu seems a lot more usable than last time I used it [09:22] then it get icons as you type, pretty nice [09:22] RAOF, do you have a stacktrace? [09:23] still needs some polish [09:23] Yeah. It's on the unity bug I filed. [09:23] Let me hunt that down... [09:24] seb128: bug #737287 [09:24] Launchpad bug 737287 in unity "Compiz crashes with SIGSEGV in nux::GraphicsDisplay::GrabPointer when opening the Dash" [High,New] https://launchpad.net/bugs/737287 [09:24] njpatel, ^known issue? [09:24] Doesn't seem to happen on my i386 install, though. [09:25] seb128, nope :/ [09:25] Will get jason to look at it today [09:25] njpatel, thanks [09:25] we are out of retracers btw [09:25] I wasn't aware they wore out :) [09:26] pitti, I did uncommit and revert the commit you pointed yesterday in the i386 retracer, no dime [09:26] seb128: :( it was a guess [09:26] seb128: good to have Alt menu keys back [09:27] not that they would actually _work_, but at least they are being displayed now when pressing alt :) [09:27] RAOF, they broke yesterday, they have an uncommon setup and didn't like the recent libstd changes in natty [09:27] pitti, btw thanks for the apport dupe sig! I haven't implemented it yet but plan to soon [09:27] pitti, oh, I didn't even notice, I never user menus [09:27] bryceh: :) I think I'll also blog about it [09:27] good idea [09:28] is launchpad working for others? [09:28] it's still spinning on RAOF's bug there but not loading it [09:29] seb128: ah, it only doesn't work in terminals, but in other programs (except firefox) [09:29] but that nicely solves the focus-follows-mouse problem [09:31] and mumble is back in the panel \o/ [09:32] bah [09:32] u1 didn't stop spamming though [09:32] that's annoying [09:32] how do you mean? [09:34] pitti, notify-osd bubbles telling that the connection to the server was dropped [09:34] then sometimes restored [09:34] but I get like a bubble every minute [09:34] it's from the sync service [09:34] yeah, I get that too [09:35] hi, could someone change bug 514401 back to Triaged? It was marked as Fix Released manually by mistake [09:35] Launchpad bug 514401 in checkbox "Translations are not loaded for the test descriptions in Checkbox" [High,Fix released] https://launchpad.net/bugs/514401 [09:36] I can't, launchpad refuses to load any bug page for me [09:37] no worries, thanks seb128. Anyone else with permissions to change bug status on src packages? [09:37] dpm: checkbox(ubuntu) right? done.. [09:38] awesome, thanks vish [09:38] np.. :) === ogra is now known as Guest78603 === Guest78603 is now known as ogra_ [10:46] pitti, do you work on the retracers issue or not? I can have a look if you don't, though I'm not sure what to do or check next [10:47] seb128: not right now, currently figuring out an input_id failure; I can look at it later on, though [10:47] pitti, ok, I will let it for you then [10:48] pitti, if you think you will not get to it let me know in the middle of the afternoon, I will give it a try [10:48] seb128: ok, thanks [10:48] thanks [10:48] ah, I also need to prepare release meeting [10:48] seb128: so if you want to try now.. [10:48] seb128: I think it might be safer to copy the auth file into the chroot instead of symlinking it [10:49] pitti, do you have any suggestion on where to start? [10:49] I will try to manually log in an do a retrace by hand [10:49] that should be a relatively easy patch in apport-chroot [10:49] just to say [10:49] see [10:49] seb128: the place you reverted yesterday -- i. e. where it currently calls os.symlink on the auth file, you could try a shutil.copy() [10:50] pitti, let me try by hand in the chroot with a symlink and with a file copy [10:50] just to see how that goes [10:50] seb128: calling apport-chroot --auth to log in worked for me yesterday [10:50] it did? [10:50] hum [10:50] yes [10:50] how is the retracing job different? [10:50] just not with the crash-digger [10:50] seb128: I don't know really :/ [10:50] oh [10:50] seb128: the symlinking is actually a bit of a hack, as it points "out of" the fakechroot [10:51] I'm not actually sure why I did that in the first place [10:52] pitti, ok, thanks, I've enough pointers to have a look I think, I will let you know how it goes [10:52] * pitti hugs seb128, good luck! [10:53] seb128: btw, the debugging code is still in chroots/natty (for amd64) [10:53] apport-retrace will print the path and contents of its --auth argument [10:53] * seb128 hugs pitti [10:53] pitti, ok === chaotic_ is now known as chaotic [11:19] is launchpad slow slow slow for others as well today? [11:20] seb128, seems a bt slower for me, yes [11:20] No, just slow slow. [11:20] seb128: seems fairly normal to me [11:20] ok [11:20] it takes me like 3 tries and 15 seconds of waiting to load a bug [11:20] where usually it's a few seconds [11:30] pitti: ping [11:30] hello Sweetshark === MacSlow is now known as MacSlow|lunch [12:16] dpm: the giraffe bug has been fixed :-D [12:17] kamstrup, \o/ thanks a lot! [12:18] dpm: and i may have nice surprise for you in an hour or so... maybe... i hope so at least... [12:18] * dpm likes surprises [12:19] excellent [12:36] dpm: sorry, have to let you down :-/ I thought I had figured out how to make nice gtk-doc from Vala, but, well, it "works" but it's definitely *not* a pretty sight :-) [12:36] dpm: so still giraffe it is [12:38] kamstrup, no worries, you're still on my hero of the day list, nevertheless :) === MacSlow|lunch is now known as MacSlow === chaotic_ is now known as chaotic [13:19] aha! take that, eclipse menus! [13:21] mterry, oh, you got those to work? [13:21] mterry, hey btw ;-) [13:21] seb128, I got them to mostly work. Which is 80% of the battle [13:22] mterry, is that application side or indicator side? [13:22] (just curious) [13:23] seb128, indicator side. We have this trick we do where we fake an activation on toplevel menu items because some apps lazily calculate menus based on activation signals. But eclipse was actually lazily calculating menus on *show* signals. So faking a show gets the menus to appear [13:24] But now it seems there are other issues with toplevel entries coming and going [13:24] oh ok [13:24] seems you got the bottom of it and are back to "standard" menu updates issues ;-) [13:24] :) [13:25] * mterry never wants to have to figure out eclipse code again. It's so big and fragmented === njpatel is now known as njpatel_ [13:42] * bcurtiswx_ waves to room and takes more aspirin for this migraine :( [13:51] kenvandine, are the current patches for empathy in natty GTK3 compatible ? [13:52] bcurtiswx_, not sure [13:52] i hope so :) [13:53] kenvandine, im building 2.91.91.1 for GNOME3 PPA but it's just from source. I plan on bringing over the patches from current natty and trying to get them to push OK and hope empathy doesn't complain about GTK3 and GTK2 being in the same codebase [13:55] wow, i haven't tried building on natty yet until now. holy cow thats much faster [13:58] i've always built natty packages with pbuilder on a maverick machine [14:03] dbarth__, seb128: just showed unity to my wife; she really liked it! I need to install natty on her laptop now :) [14:06] pitti, great [14:06] ok, so so launchpad issues was on my side, restarting the modem solved it === seb128_ is now known as seb128 [14:09] pitti, is the rt meeting one hour early or is my calendar off? [14:10] seb128: it's early, as the US is already on DST [14:10] ok [14:10] so it's not sitcked on an UTC time [14:11] sticked [14:12] pitti, so doing a shutils.copy doesn't solve the retracer issue [14:12] pitti, do you know why you debug modif don't show in a bzr diff btw? [14:12] seb128: it's not in a bzr branch, I only made it in the installed files in the fakechroot tarball [14:12] oh ok [14:13] stil doing the no such file or directory [14:13] crap [14:13] I don't get why the auth file is working in a manual logging though [14:13] seb128: does the debug info show "/tmp/auth" now instead of /tmp/tmpdir/tmp/auth? and still no file contents? [14:13] or what is different between those [14:13] IOError: [Errno 2] No such file or directory: '/tmp/tmpBET1Mu/tmp/auth' [14:14] pitti, no [14:14] ah, so perhaps it's not the link itself, but the --auth argument is passed wrongly [14:14] while we had the link, cat'ing it in the shell worked fine [14:15] seb128: hang on [14:16] seb128: the symlinking is done in two paces, in command_retrace() and command_login(); I think back then I only patched command_login() [14:16] _that_ would explain the difference! [14:16] pitti, let me try that [14:16] seb128: I bet if command_retrace will also use the abspath() thing, it'll work [14:16] bah, silly me [14:17] hm, no didrocks today? [14:18] mterry: should we postpone the quickly bits to o? seems this is still blocked on LP? [14:18] pitti, he's having a day off today [14:18] (well deserved) [14:19] pitti, oh yeah, it's unlikely they'll add the bits to LP in time for me to react to it [14:19] i'll postpone [14:19] mterry: ok, thanks; so at the start of oneiric we should the re-target the entire spec [14:19] and change the POSTPONED into TODO again [14:21] pitti, the command_retrace() thing, should that be patched in the checkout or in the tar.gz? [14:21] seb128: in the checkout [14:21] pitti, ok, so doesn't work [14:22] seb128: apport-chroot is run "outside" and is the thing that unpacks the chroot tarball and runs stuff in int [14:22] 'it' [14:22] pitti: :) sweet [14:23] seb128: hm, looking at your diff now, looks correct [14:23] seb128: does that one work with copy? [14:23] seb128: conversely, does the copy one work for login? [14:23] seb128: ah [14:23] seb128: ignore me [14:24] seb128: try replacing [14:24] apport_retrace_argv += ['--auth', chroot_auth] [14:24] with [14:24] pitti, trying with shutils.copy [14:24] apport_retrace_argv += ['--auth', '/tmp/auth'] [14:24] chroot_auth is the full "outside" path [14:24] (the abspath change is still correct, though) [14:25] I guess it was a mere coincidence that this worked under lucid [14:26] pitti, it's weird that those behaviour changed with the distro version [14:27] seb128: I guess it was fixed in fakechroot [14:27] a path like /tmp/tmp2304823432/tmp/auth isn't actually supposed to be valid within the fakechroot [14:28] it's just /tmp/auth there [14:30] rodrigo_: note that you seem to have forgotten the patch attachment in https://bugzilla.gnome.org/show_bug.cgi?id=644560 [14:31] Gnome bug 644560 in Gal "Can't set an account password" [Normal,Needinfo] [14:31] pitti, \o/ [14:31] ---- apport-retrace auth file: /tmp/auth [14:31] it's retracing! [14:31] yippie [14:31] pitti, can you commit to trunk? [14:31] so I just have to bzr pull [14:31] seb128: that's in the Ubuntu branch only; will do, thanks for testing! [14:31] yw [14:31] hum, it crashed [14:31] seb128: does it also work with symlink? [14:32] (it ought to) [14:32] self.duplicate_db = dbapi2.connect(path, timeout=7200) [14:32] sqlite3.OperationalError: unable to open database file [14:32] pitti, I guess you need the same fix for apport_retrace_argv += ['--duplicate-db', chroot_dupdb] [14:33] seb128: yep, committing [14:34] seb128: give me a sec to commit, push, pull, and clean up [14:36] seb128: ok, pulled into both apport-retracer-*/apport branches [14:36] seb128: mind to restart and check? [14:36] pitti, no, just that I discussed it with upstream and the fix was a bit more complicated, so working on it [14:36] pitti, trying [14:37] rodrigo_: ah, ok; the comment said "Attaching a fix.." [14:37] seb128: (sorry, need to finish the release team prep; spent too much time on fixing input_id :/) [14:37] pitti, nothing to be sorry about, thanks for spending time on that [14:37] ---- apport-retrace auth file: /tmp/auth [14:37] pitti, so it's starting retracing [14:38] let's see if the db thing is fixed as well [14:38] Duplicate check negative [14:38] New attachments uploaded to crash database #737294 [14:38] \o/ [14:38] pitti, thanks a lot [14:38] yippie [14:38] let me update the other one and remove the locks [14:39] seb128: I already updated bzr; you migh tneed to dist-upgrade the fakechroot [14:39] (althuogh that should happen automatically) [14:39] pitti, right that should happen [14:39] I will just be lazy, remove the lock and wait the next run [14:39] I will watch the log and debug if needed === ogra_ is now known as ogra [15:12] why don't we package telepathy-qt4? j/w [15:13] bcurtiswx_, you are welcome to work on that [15:13] pitti, retracer is retracing happily \o/ [15:13] seb128: cool! [15:14] since we don't package it, and it's just as easily sync'd, what needs to be done? just a sync request? [15:16] i see it's in debian, but not since 0.3.8 [15:16] its 0.5.13 now [15:17] bcurtiswx_: yes, a sync request should be enough but you need to check if you might also need a FFe (and that it builds in Ubuntu) [15:18] geser, OK thx [15:18] bcurtiswx_, there is no new version in debian [15:19] http://packages.qa.debian.org/t/telepathy-qt4.html [15:19] seb128, yes i saw that. It's far behind the current version and it even says in there it sees a newer version [15:20] but the newest isn't in debian [15:20] well ask on #telepathy I guess [15:20] is telepathy-qt used anywhere? [15:20] seb128, lol i was already typing my message.. [15:24] bcurtiswx_: ? we do package telepathy-qt4 [15:25] Riddell, not at the moment. I'm on the telepathy mailing list and I get all their package updates. Wondering really what telepathy-qt4 is and if it is worth even packaging. My first assumption would be that since we don't package it now, it probably doesn't make a huge difference not having it [15:30] is we don't need qt support for telepathy, then we don't need tp-qt4 [15:30] s/is/if [15:32] oh gosh, nvm [15:32] im gonna go hide in a corner for a little while [15:46] for future reference, don't go trying to figure things out mentally when you have a migraine [15:48] b'ah, something has changed in natty which breaks firefox builds [15:56] chrisccoulson: multiarch? [15:57] pitti - micahg suggested that, but it doesn't look like a multiarch problem [15:57] i guess it could be though [15:57] i'm just looking at what packages to downgrade [15:57] but the build system looks totally screwed [15:58] where are the daily isos for natty? [15:59] ah, found it in firefox hstory! === tkamppeter_ is now known as tkamppeter [16:15] ok, tested the actual rc2 build, and that fails too [16:16] :( [16:46] pitti / micahg - oh, it is multiarch related. downgrading eglibc fixes the firefox build [16:46] now to figure out why [16:54] bcurtiswx: https://launchpad.net/ubuntu/+source/telepathy-qt4 says we do package it [16:54] bcurtiswx: we need it for the forthcoming telepathy-kde bits. it could do with being updated to the latest version [16:54] unfortunately there's no easy way to subscribe to new version notifications, it's just announced on their mailing lsit [16:54] list [17:29] mterry, bug #735635 could be a bug in your ntp backport, well rather in the error handling code from it [17:29] Launchpad bug 735635 in gnome-settings-daemon "gsd-datetime-mechanism assert failure: *** glibc detected *** /usr/lib/gnome-settings-daemon/gsd-datetime-mechanism: double free or corruption (out): 0x00404c70 ***" [Medium,New] https://launchpad.net/bugs/735635 [17:30] the crash is in gsd_datetime_mechanism_set_using_ntp() g_set_error() [17:35] seb128, looking [17:38] Sweetshark: I updated the oo.o -> libo stuff in language-selector now, FYI [17:42] pitti: ah, cool [17:42] * Sweetshark currently merges 3.3.2-1 from debian [17:42] ... and adds lpi integration again. [17:43] seb128, solved, thanks for letting me know. stupid error on my part [17:44] (basically completely rewritten that is) [17:44] mterry, thanks, I figured it was probably an easy one that's why I pinged ;-) [17:53] hum [17:53] bug #737651 [17:53] Launchpad bug 737651 in gdm "gdm fails to load /usr/bin/X after update to Xorg 1.10.0-0ubuntu1" [Undecided,New] https://launchpad.net/bugs/737651 [17:53] bryceh, ^ it might be worth watching, not sure if that could be a breakage of today's update === evilvish is now known as vish === MacSlow is now known as MacSlow|errand [18:10] good bye everyone, have a nice weekend! [18:11] pitti, thanks, you as well! === njpatel is now known as njpatel|away [19:02] seb128, thanks; seems to be an isolated report so far (and I haven't reproduced it myself), but will keep an eye out for dupes [19:04] pitti, is there a way to do something like the dupe signature for bugs like 737453 and 737636? Or else some way to get jockey/apt/apport to let the user know fglrx/nvidia isn't available if they upgrade while it's ABI broken? [19:04] pitti, we get ~1-2 bugs a day like these, so it'd cut down on a goodly bit of bug noise [19:05] hmmm, the firefox build failure is a make bug! [19:05] (because of multiarch) === MacSlow|errand is now known as MacSlow === ogra is now known as Guest95250 [20:56] pitti, ok DuplicateSignature implemented in -intel gpu lockup apport hook... now to watch if it works :-) [22:28] i can't believe how long it takes natty to boot on my laptop atm === Guest95250 is now known as ogra