[00:25] stgraber: how do I check the test results? [00:33] BenC: no idea, in my case the build died right after the failure. If you actually get debuild to succeed, then it didn't fail for you [00:35] stgraber: it failed in the tests, but I'm trying to find the output of those failures and it isn't very easy [00:36] how do i make an "applet" that runs in the ubuntu "system-tray" ? [00:38] bretth: YOu want to create an indicator. Best you ask on #ubuntu-app-devel, I am pretty sure someone in there can point you to indicator documentation. [00:38] TheMuso, thanks! [00:41] Is there an easier way to add a custom .udeb to an initrd.gz than rebuilding the debian-installer? === cpg is now known as cpg|away === Ursinha` is now known as Ursinha === Ursinha is now known as Guest53271 === rsalveti` is now known as rsalveti === cpg|away is now known as cpg === Zic is now known as Guest2281 === cpg is now known as Guest92579 [02:19] what's the best way from python for my script to know that it is running in Ubuntu with Unity? [02:20] (my script needs to support multiple desktops, Gnome3, Cinnamon, etc) [02:21] i was thinking that "try: import appindicator" is one way, but kind of looks like a hack. [02:29] bretth: I believe jockey does something similar to check environment. Check the jockey source package from Ubuntu, quantal should do, but precise would be a better option, as jockey has changed somewhat so far as I understand things... [02:29] bretth: Jockey also has multiple frontends for GTK and KDE. [02:31] @pilot in === udevbot changed the topic of #ubuntu-devel to: Quantal Quetzal A2 released! | Archive: open | Dev' of Ubuntu (not support or app devel) | build failures -> http://bit.ly/HaWdtw | #ubuntu for support and general discussion for hardy -> precise | #ubuntu-app-devel for app development on Ubuntu http://wiki.ubuntu.com/UbuntuDevelopment | See #ubuntu-bugs for http://bit.ly/lv8soi | Patch Pilots: RAOF [02:32] TheMuso, thanks, another question, how long will Ubuntu support pygtk (gtk2)? is it safe to write my app using pygtk? [02:52] bretth: If you are supporting precise or later, I strongly suggest you use Gtk2 via GObject introspection. [02:52] Or even better, use GTK3. [02:52] i need to also support Ubuntu 10.10 [02:52] Pygtk2 is not supported upstream really, so far as I know, if anything, they want it gone. [02:53] Ah ok, then I think pygtk2 is your best bet then. [02:53] bretth: Ubuntu 10.10 is no longer supported, why would you want to support it? [02:53] micahg, its not my decision [02:54] bretth: 10.04 is supported on the desktop for another 9 months [02:54] 10.04 is Unity? [02:55] bretth: no, it's pre unity, but even in precise, you don't have to use Unity, there are many other desktop environments [02:56] micahg, seems that must use Unity for appindicator. [02:56] gnome2 had indicator-applet, iirc [02:56] And indicators are used in other DEs as well. [02:56] bretth: no, I don't think that's true, there are indicators in xubntu [02:56] They're there by default on Xubuntu. [02:56] *Xubuntu [02:57] Kubuntu may also do something with them, but I don't remember. [02:57] oh good, that makes it easier to support Xubuntu. [02:57] the KDE panel tray supports anything conforming to the indicator spec [02:58] i already have a applet designed for KDE [02:58] oh yeah, you were in #kde-devel last night I think [02:58] yep, that was me. [02:59] so is Wayland making it into the next Ubuntu? [03:00] I don't think that'll happen for a while. *Maybe* there will be a tech demo available for installation in the archives for the next Ubuntu. [03:00] i have python wrappers for wayland working [03:00] works with Python2, Python3, and PyPy [03:00] http://code.google.com/p/rpythonic/source/browse/examples/wayland-server-eventloop.py [03:01] http://code.google.com/p/rpythonic/source/browse/examples/wayland_server/__init__.py [03:01] !info weston quantal [03:01] weston (source: weston): reference implementation of a wayland compositor. In component universe, is optional. Version 0.85.0-1build1 (quantal), package size 178 kB, installed size 437 kB (Only available for linux-any) === Guest92579 is now known as cpg [03:29] JontheEchidna, does it work with gtk now? [03:33] scientes, i saw some screen shots of gtk on wayland. [03:33] yes yes i've used it myself [03:33] but i mean in ubuntu [03:33] without building anything [04:21] cjwatson: ping [04:56] Good morning [05:01] Good morning pitti. [05:02] pitti: Just so someone else other than me knows the status: We've had Bug #1024016 and a fair number of duplicates pile up yesterday and tonight. I think I just uploaded the final bit of fixing it. [05:02] Launchpad bug 1024016 in python3-defaults (Ubuntu) "package python3 3.2.3-2 failed to install/upgrade: subprocess installed post-installation script returned error exit status 4" [Critical,Fix released] https://launchpad.net/bugs/1024016 [05:04] Made the publisher run by a few seconds, so we should see goodness in half an hour. [05:07] I grabbed the .debs from LP and confirmed it's really, really fixed now. [05:07] ah, seems I just ran into this during daily dist-upgrade; thanks! [05:08] You can either grab the new versions of whatever you have installed from python3-defaults or wait a bit I guess. [07:02] good morning === zyga-afk is now known as zyga === smb` is now known as smb === ubuntu is now known as Nathan_S === henrix_ is now known as henrix === tkamppeter_ is now known as tkamppeter [08:31] @pilot out === udevbot changed the topic of #ubuntu-devel to: Quantal Quetzal A2 released! | Archive: open | Dev' of Ubuntu (not support or app devel) | build failures -> http://bit.ly/HaWdtw | #ubuntu for support and general discussion for hardy -> precise | #ubuntu-app-devel for app development on Ubuntu http://wiki.ubuntu.com/UbuntuDevelopment | See #ubuntu-bugs for http://bit.ly/lv8soi | Patch Pilots: [08:34] * dholbach hugs RAOF [08:34] hum, is anyone else having an issue with python3-minimal installation in quantal? [08:34] Morning dholbach :) [08:34] hey :) [08:35] dholbach: That should be fixed in python3 3.2.3-3ubuntu1 === ubuntu is now known as Guest86821 [08:35] You're seeing bug #1024016 [08:35] Launchpad bug 1024016 in python3-defaults (Ubuntu) "package python3 3.2.3-2 failed to install/upgrade: subprocess installed post-installation script returned error exit status 4" [Critical,Fix released] https://launchpad.net/bugs/1024016 [08:36] ok, manually downloading the packages and installing them with dpkg -i fixed it [08:36] but apt-get install -f didn't === Guest86821 is now known as Nathan_S [08:45] any perl guru's around? I need some help with bug 1022385 - seen this a few times now and not really sure what it points to [08:45] Launchpad bug 1022385 in samba (Ubuntu) "package samba 2:3.6.3-2ubuntu2.3 failed to install/upgrade: subprocess installed post-installation script returned error exit status 128" [Undecided,New] https://launchpad.net/bugs/1022385 [08:47] the exit code may be unrelated to the noise from debconf, since none of that noise ought to be fatal [08:47] 128 is an unlikely exit code for a debconf-related problem; but you can try with DEBCONF_DEBUG=developer to establish that more certainly [08:48] I think that noise basically means that the user disconnected from debconf, maybe by closing a dialog box or something [08:50] Thanks. I'd left the bug there because I wondered if anyone else knew what was going on [08:51] rbasak, I found another 128 code with bacula with similar symptoms hence my question - cjwatson - OK I'll get the reported to try that then [08:51] ta [08:51] cjwatson, does exit code 128 have any specific meaning? [08:51] (and where would I look for that sort of thing) [08:53] There's no single place you can look. Exit codes are very application-specific. [08:56] I can't think of any specific meaning for 128 that might have broad applicability. === fenris is now known as Guest73296 === cpg is now known as cpg|away === Guest2281 is now known as Zic [09:05] hggdh: Don't suppose you had any luck getting me that debug output? [09:06] hggdh: There seems to be some in https://jenkins.qa.ubuntu.com/view/Quantal/view/ISO%20Testing%20Dashboard/job/quantal-desktop-i386_default/69/, but it doesn't contain the bits I expected ... [09:30] cjwatson: I am executing another job, on a particular node and will connect to the VM whenever it's done installing [09:30] cjwatson: I should be able to get you some logs out of it [09:31] OK, let's hope it actually fails [09:31] cjwatson: indeed [09:31] Currently uploading ubiquity 2.11.14 with an utter-guesswork possible fix (I think it's a long shot but you never know) [09:32] cjwatson: will I be able to connect to the VM if it never finishes installing? [09:32] I don't know [09:32] That all sounds rather specific to your test environment, and I'm not familiar with it [09:33] cjwatson: I am learnin as I go, but the installer just said that it installed successfully :/ [09:35] Try again [09:35] ack [09:36] It's racy - even on jenkins, which seems to reproduce this more frequently than any other environment (maybe just because it does lots of runs), it only happens sometimes [09:36] ok [09:36] I think only one of last night's runs failed [09:36] https://jenkins.qa.ubuntu.com/view/Quantal/view/ISO%20Testing%20Dashboard/job/quantal-desktop-i386_encryptedhome/46/artifact/46/test-results/d-i-debug.log exists but is empty :-( [09:37] cjwatson: do you know roughly at which point in the installation this happens? [09:37] hggdh: Whatever you did in https://jenkins.qa.ubuntu.com/view/Quantal/view/ISO%20Testing%20Dashboard/job/quantal-desktop-i386_encryptedhome/46/ seems to have made the d-i-syslog useless too [09:37] gema: "Configuring hardware" [09:37] Fairly near the end [09:37] hggdh: That seems to be more like casper.log or something ... [09:38] cjwatson: I tried on one of my test machines the other day and it got stuck on the camera screen to never come back [09:38] gema: Whatever that is (a gstreamer bug, I guess), it's unrelated [09:39] cjwatson: ack [09:55] didrocks, can you join #ubuntu-arb again? :-D [09:57] dholbach: yeah, coming. Should really adjust my autojoin :) [09:57] :) [10:12] i have a dependency problem with python3 on quantal: http://paste.ubuntu.com/1089585/ [10:14] could somebody set https://code.launchpad.net/~lars.duesing/ubuntu/precise/aiccu/aiccu-upstart-server-patch-precise/+merge/111610 as merged? [10:15] the fix got uploaded but the merge request was targetting precise and not precise-proposed [10:18] seb128: done [10:21] pitti, danke === ricotz_ is now known as ricotz === MacSlow is now known as MacSlow|lunch [11:56] cjwatson: all I did was pass to the kernel debug-ubiquity [11:58] cjwatson: and the debug log seems to be empty. I tried /var/log/debug as I was told, and /var/log/installer/debug, as the wiki says [11:58] there, StacktraceSource.txt is being produced again: http://launchpadlibrarian.net/110026354/StacktraceSource.txt [11:59] cjwatson, green jenkins on both amd64 and i386, either we got lucky or you fixed the bug ;-) [12:04] seb128: we got lucky [12:04] I suspect [12:04] seb128: oh, do you mean with .14? [12:06] seb128: there've not been any tests with the latest build yet, so really can't say anything [12:06] cjwatson, no, 13.1 for both [12:06] hggdh: I did tell you /var/log/installer/debug, not /var/log/debug (which would be wrong) [12:06] cjwatson, I though .1 was your recent upload [12:07] cjwatson, .14 is tomorrow ;-) [12:07] cjwatson, well, anyway, let's see === _salem is now known as salem_ [12:31] seb128: retracers (hopefully) fixed again, after the two recent cron mails [12:31] pitti, thanks [12:31] pitti, what was the issue on those? === zyga is now known as zyga-food [12:32] seb128: the first one was a bug when a source file had non-ascii characters, the retracer is running under C locale [12:32] seb128: the second just me picking the wrong cache dir when running apport-retrace manually [12:32] did I say how much I hate encodings? ;-) [12:32] seb128: it got exposed because StacktraceSource.txt is back on now [12:33] nice to see those back, thanks for that [12:33] seb128: yeah; one of these years, C and all other non-UTF-8 encodings just belong abolished.. [12:33] it's handy to have them [12:33] seb128: yeah, everyone should speak ASCII-only languages [12:33] (shame English doesn't qualify ...) [12:33] english is not ascii only? [12:33] "naïve" [12:34] yeah, http://launchpadlibrarian.net/110026354/StacktraceSource.txt looks quite nice [12:34] you stole that to us, didn't you? ;-) [12:34] "fête" - various odds and ends like that [12:34] ("from"?) [12:34] and yeah, mostly French imports :) [12:35] actually these days it's mostly legit to drop the diacritics but anyway [12:35] cjwatson: don't forget about the ☺ [12:36] * pitti imagines Kanji or Mandarin in ASCII art [12:40] pitti, Ṳßυᾔ⊥υ [12:40] pitti, http://fsymbols.com/generators/encool/ does that sort of stuff ;-) [12:41] you type ascii chars and it gives you a fancy similar looking one made from unicode chars [12:41] ¡ʎuunɟ sʞooן ʇɐɥʇ [12:41] ∂ε♭☤αη [12:41] oops, /me turns his monitor back up === zyga-food is now known as zyga [12:41] lol [12:41] . o O { clearly Friday afternoon.. } [12:41] pitti, how did you do that? ;-) [12:42] seb128: I turned my monitor upside down [12:42] hehe [12:42] nah, http://www.sevenwires.com/play/UpsideDownLetters.html [12:42] ok, we just traded fun websites then :p [12:42] ⊥♄@☂❝ṧ ηḯ¢ε [12:42] it's friday indeed ;-) [13:04] mpt: looking at the several week view of "average crashes per day." Rather unsurprisingly, theres a bump on the weekends. [13:05] didrocks: ping [13:05] zul: pong [13:05] ev, which isn't what we want ... You shouldn't be able to improve the stats by making Ubuntu so that people don't want to use it [13:05] didrocks: can you have a look at the python-quantumclient MIR (LP: #1021822) [13:07] zul: what is depending on it? [13:07] didrocks: new nova build depends [13:07] zul: can you please state that in the MIR? [13:07] mpt: hmm? In that because it doesn't factor in the total usage it's skewed on the weekends? Thus if we make people use Ubuntu for shorter periods of times the graph will look better, but the experience will still be poor? [13:08] didrocks: done [13:08] zul: all build-deps are not in main [13:09] zul: python-nosexcover is not for instance [13:09] didrocks: crap [13:09] zul: can you please check it again? the stenza in the MIR is not just for copy and paste :) [13:10] zul: please reassign once sorted out :) [13:12] barry: is there a bug open for the uninstallable python 3 in quantal? [13:13] can't be entirely uninstallable, livefs builds haven't broken === yofel_ is now known as yofel [13:13] (I haven't looked at it at all mind you and won't, enough on my plate) [13:14] @pilot in === udevbot changed the topic of #ubuntu-devel to: Quantal Quetzal A2 released! | Archive: open | Dev' of Ubuntu (not support or app devel) | build failures -> http://bit.ly/HaWdtw | #ubuntu for support and general discussion for hardy -> precise | #ubuntu-app-devel for app development on Ubuntu http://wiki.ubuntu.com/UbuntuDevelopment | See #ubuntu-bugs for http://bit.ly/lv8soi | Patch Pilots: ogasawara [13:16] dobey: did you figure out your python3 install issue? is there a bug open? [13:17] mdeslaur: i haven't figured it out, no. there is a bug with many dupes afaik. let me see if i can find the # [13:17] mdeslaur: https://bugs.launchpad.net/ubuntu/+source/python3-defaults/+bug/1024016 i think [13:17] Launchpad bug 1024016 in python3-defaults (Ubuntu) "package python3 3.2.3-2 failed to install/upgrade: subprocess installed post-installation script returned error exit status 4" [Critical,Fix released] [13:19] dobey: ah, yes, that's the one...thanks [13:21] dobey: ah, manually upgrading python3 as suggested at the end just worked for me [13:25] yeah i just had to manually install the 3ubuntu1 version, then do -f install [13:32] mvo: hey Michael, how are you? [13:32] mvo, aptdaemon SRU veri^W forget about that you did it :p [13:32] ;-) [13:32] hey pitti, good, thanks! how are you (phonecall though right now) [13:33] mvo: good, thanks! currently investigating some auto-upgrade-tester failures === MacSlow|lunch is now known as MacSlow === rsalveti` is now known as rsalveti [13:37] ogra_: hi there, roadmr and I need your assistance to close bug #245488 by removing the ol' hwdb-client package from universe :) [13:37] Launchpad bug 245488 in hwtest (Ubuntu) "please add proper replaces for hwdb-client to hwtest " [Medium,Invalid] https://launchpad.net/bugs/245488 [13:37] pitti: aha, precise->quantal? or lucid->precise? [13:37] mvo: lucid->precise, but both really [13:37] cr3, yeah, i saw the bugmail, why the heck is that still there ? [13:38] mvo: WDYT about http://paste.ubuntu.com/1089849/ ? [13:38] ogra_: no clue but, instead of just marking the bug as invalid, we'd like to take this opportunity to clean the archive a bit [13:38] cr3, and that bug is obviously invalid now [13:38] ah [13:39] ogra_: I'm using the bug as a nagging reminder to do things right :) [13:39] https://wiki.ubuntu.com/UbuntuDevelopment/PackageArchive#Removing_Packages [13:39] mvo: I added a reproducer and analysis to bug 1024346 and verified that it works fine now [13:39] Launchpad bug 1024346 in auto-upgrade-testing (Ubuntu) "python import test fails on modules which need $DISPLAY" [High,In progress] https://launchpad.net/bugs/1024346 [13:41] ogra_, roadmr: 23 bugs for hwdb-client in ubuntu: https://bugs.launchpad.net/ubuntu/+source/hwdb-client [13:41] ogra_: thanks for the link, we'll follow the process [13:41] cr3, just doing so === fabbione is now known as fabbione_vac [13:42] ogra_: sweet, thanks! [13:42] cr3: does checkbox still have any file overlaps with hwdb-client? [13:42] cr3: I got rid of all the hwtest bugs, didn't touch hwdb-client [13:44] cjwatson: nope, hwdb-client was there such a long time ago, it was even renamed to hwtest before being finally renamed to checkbox [13:44] right [13:44] ogra_: the good ol' days :) [13:44] cr3: since it is still in the archive, if there were any file overlaps then they can't be excused by intermediate renamings [13:44] so is the answer yes or no? :) [13:45] cjwatson, i changed the bug above into a removal bug for hwdb.-client [13:45] I know, but that doesn't answer my question either [13:45] roadmr: when did we remove the transitional package for hwtest -> checkbox? is it still in precise? [13:45] cr3: nope, last release it shows for is oneiric [13:46] how could there be file overlaps if one is in universe and the other is in main [13:46] ogra_: that said, I'm confused, hwdb-client is not in quantal [13:46] even shipped on the cd [13:46] ogra_: overlaps in their *contents* [13:46] cr3: removed since 2012-02-01 [13:46] universe/main> totally irrelevant [13:46] right, I removed hwdb-client from intrepid in 2008. I don't know why you're asking for it to be removed now :) [13:47] https://launchpad.net/ubuntu/+source/hwdb-client/+publishinghistory [13:47] * ogra_ doesnt get that, if checkbox works standalone without hwdb-client installed, where should be file overlaps ? [13:47] sigh [13:47] you know what replaces: is for, right? [13:47] yes [13:47] among other things, it's for when a package ships a file previously shipped by another package [13:47] I looked for hwdb-client and I found it in universe: http://archive.ubuntu.com/ubuntu/pool/universe/h/hwdb-client/ [13:47] yes [13:47] cjwatson: I just checked the packages in the archive and there are no overlaps with any files in checkbox [13:47] roadmr, thats old stuff [13:48] if checkbox had shipped a file that hwdb-client previously shipped, then it ought to have had a Replaces header, despite hwtest coming in between [13:48] cjwatson, i still dont get how that would be relevant for the removal of hwdb-client [13:48] ogra_: it might've been possible that both hwdb-client and checkbox installed /usr/bin/foo, for example. I think that's what we're looking for [13:48] it's not particularly, it's just that sometimes people think that removing the old package is good enough in such cases, and it isn't [13:48] oh, foo [13:48] I was checking that due diligence had been done [13:48] ah, ok [13:48] cjwatson: hwtest, that followed hwdb-client, had a Replaces header [13:49] cjwatson: err, checkbox has a Replaces header for hwtest, that followed hwdb-client [13:49] understood but not relevant :) [13:49] cjwatson: were you talking (when you said the ubiquity syslog was useless) about the 'not copying..." etc? [13:49] in general we cannot assume that users have traced through every step in our package renamings [13:49] anyway, this is moot, because hwdb-client was gone in intrepid and you've said that there are no file overlaps [13:49] hggdh: in the URL I gave, it's not even a ubiquity syslog. it stops at the end of casper. [13:50] mpt: ah, excellent. That's much better than dealing with it off a subpage [13:50] gema, mvo: do you know whether the jenkins upgrade tests use the branch or the package for auto-upgrade-testing? [13:51] gema, mvo: i. e. I hope we don't need to SRU the latest changes just to fix the tests [13:51] pitti: I don't really know [13:52] cjwatson: yes, this build failed -- we got an error relating to hardware, and the install got stuck on a pop-up, and was eventually canceled on timeout [13:56] pitti: I'm pretty sure it uses the package [13:56] mvo: gah [13:56] mvo: anyway, do you think the xvfb-run patch is appropriate? [13:57] hggdh: ok [14:13] ScottK: thanks for fixing the py3 mess! [14:14] pitti: yeah, I think that is a excellent idea [14:14] mvo: ok, thanks; committing that and uplaoding to quantal, so that at least teh precise->quantal tests should have this fixed [14:14] pitti: Thanks. It was my sync that caused it, so I think it was mine to fix. [14:15] pitti: cool, thanks [15:04] seb128, are you aware of compiz constantly restarting when using Libreoffice? [15:04] jono, no [15:05] jono, seems like a segfault, what Ubuntu version, is apport triggering? [15:05] seb128, I have been working on a presentation and it keeps restarting for me [15:05] apport isnt triggering [15:05] I am on Quantal [15:05] jono, tail /var/log/apport.log? [15:06] seb128, no output [15:07] jono, dmesg | grep compiz ? [15:09] seb128, [73664.275452] compiz[1828]: segfault at 64 ip b45b0677 sp bfac8f90 error 4 in libunityshell.so (deleted)[b4401000+307000] [15:10] jono, there you go [15:10] jono, sudo service apport start force_start=1 [15:10] jono, get the issue [15:10] jono, report the bug [15:10] start: Job is already running: apport [15:11] jono, ok, not sure why apport is not working then [15:11] jono, nothing in /var/crash ? [15:12] seb128, bunch of things in there [15:12] jono, compiz one? [15:13] jono, you might want to wipe them all, trigger the bug and see if a compiz one gets written [15:13] doko, hey, is https://bugs.launchpad.net/ubuntu/+source/gdb/+bug/1006860 good to upload or do you want to review,handle it? [15:13] Launchpad bug 1006860 in gdb (Ubuntu) "gdb crashes when loading core files (in is_ctor_or_dtor)" [High,Confirmed] [15:15] seb128, won't have time for it. a NULL check doesn't sound bad. please could you keep the report open? [15:15] doko, is it fine if I upload? it's blocking the dx guys in some of the debugging they do [15:29] chrisccoulson, ping [15:30] doko, probably leave some context, he's off for 2 weeks on paternity leave [15:30] doko, he might look at IRC every now and then but is not working === allee_ is now known as allee [15:46] seb128, just retriggered the bug, no apport crash [15:46] seb128, sorry for the delay, was on a call [15:46] jono, nothing in /var/log/apport.log? [15:47] seb128, nope [15:48] jono, do you have a ~/.apport-ignore.xml? [15:48] seb128, yep [15:48] jono, does it contain something saying to ignore compiz issues? [15:48] [15:48] nope [15:49] :-( [15:49] you would need pitti then, not sure why apport doesn't work [15:49] but he called it a week already I think [15:49] thanks seb128 [15:49] jono, I get you are down to "use gdb" [15:49] if you want a stacktrace [15:50] I will see if this is still happening [15:50] seb128, if you open LO and edit a presentation, does the bug occur for you? [15:57] jono, wb, do you think I'm crazy enough to run quantal yet? :p [15:58] seb128, are you serious? lol [15:58] seb128, I thought you would need to be running it [15:58] :-) [15:58] jono, I'm part of the LTS .1 team in fact so for once I'm staying on stable for some time ;-) [15:59] jono: I see no crash in that case. However, the top unity panel becomes transparent, which is an interesting case on itself :) [15:59] when you double click to select something [15:59] well, not everytime [15:59] but common enough [15:59] jono, it would help to get a stacktrace of the issue, do you know how to do that? [15:59] seb128, nope [16:00] I just rebooted and I can't reproduce [16:00] hum, k [16:00] maybe this was fixed in an update [16:00] just trying to trigger it now [16:00] could be [16:00] didrocks, hah! [16:00] well, there is transparent panel to get fixed :) [16:00] aha [16:00] it just happened [16:00] how do I get a trace seb128? [16:00] didrocks, why do you hate the bling? ;-) [16:00] jono: if you get something similar, try to hover the panel [16:00] see if it's a crash [16:01] or something like I experience [16:01] seb128: well, I think we just created the tag in unity :) [16:01] jono, apport is still not logging any compiz file in /var/crash even after reboot? [16:01] didrocks, I see Unity seemingly restart and LAuncher vanishes and then returns [16:01] seb128, nope [16:01] jono: so maybe it's the same bug, it's not crashing, it's just blinking [16:01] that's why you don't get a crash file :) [16:02] jono, basically, go to a vt, sudo gdb -p $(pidof compiz), on the compiz prompt: [16:02] set logging on [16:02] c [16:02] tsss [16:02] return to your session [16:02] get the bug [16:02] unity --advanced-debug in a vt [16:02] why do I work for otherwise? :) [16:02] jono, ^ what didrocks says [16:03] didrocks, does that log into a file? [16:03] didrocks, or do you need to add --log? [16:04] seb128: I'm looking, I have a fish memory on what I did :) [16:04] jono, key point is to run those commands from a vt or another session since when compiz will segfault you will not be able to access your session to go to gdb if you run it there [16:04] hmm switching VTs seems to restart compiz too [16:04] ok I ran --advanced-debug in a bt [16:04] vt [16:05] jono, good, get the bug now ;-) [16:05] so just trigger it? [16:05] yes [16:05] that will freeze your xorg session [16:05] got to the vt [16:05] type "bt" [16:05] and quit gdb [16:05] that will restore your graphical session [16:06] ok triggered [16:06] yeah, it should log [16:06] bt says "no stack" [16:06] yeah, so I'm sure you have the same thing than I have [16:06] not a crash [16:06] just the launcher/panel blinking [16:07] oh [16:07] ahhh gotcha [16:07] seb128, you should upgrade to see it ;-) [16:07] jono: want to open a bug or me to do it? [16:07] didrocks, would you as you can probably add more detail [16:07] all I can say is that it blinks [16:07] jono: sure [16:07] I will add the details of LO triggering it though [16:07] thanks didrocks! [16:08] jono, I will put quantal on my netbook, I want to play with the system compositor stuff from robert_ancell anyway ;-) [16:08] thanks for the help seb128 [16:08] jono: I will write it in French to punish the upstream team :) [16:08] didrocks, lol [16:08] jono, yw, that also explain why apport didn't work ;-) [16:08] seb128, indeed [16:08] :-) [16:08] brb [16:13] (if only launchpad would let me file a bug) [16:14] seb128: does https://bugs.launchpad.net/unity/+filebug works for you? [16:15] didrocks, it loads [16:15] not for me :/ [16:15] didrocks, do you want me to try to submit? [16:15] ok, ubuntu-bug unity FTW [16:15] seb128: that's fine, I workaround it :) [16:15] thanks! [16:15] ok [16:15] np ;-) === ubuntu is now known as Nathan_S [16:20] jono: if you want to track it, bug #1024459 [16:20]