[00:53] mlankhorst: is there a ppa anywhere that will get me DRI3 and Present? [01:17] ali1234: theres no driver support upstream yet except obscure branches that may or may not work, i doubt it [01:17] i don't need driver support at this point - present should "work" without driver support [01:18] at least in the sense it will display things on the screen. just won't be vsynced yet === Ursinha is now known as Ursinha-afk [01:19] hmm, https://launchpad.net/~canonical-x/+archive/x-staging then? [01:19] no it needs mesa git also [01:21] i guess i'll put this on hold until more stuff filters through === duflu_ is now known as duflu === Ursinha-afk is now known as Ursinha [03:18] [precise 12.04.3] trying to troubleshoot Unity greeter's 'Shutdown' being broken. tracked it down to /usr/lib/indicator-session/gtk-logout-helper hitting a dbus AccessDenied when calling ConsoleKit Stop method. can reproduce using dbus-send. how do I troubleshoot from here? [03:19] *error is actually org.freedesktop.ConsoleKit.Manager.NotPrivileged === Ursinha-afk is now known as Ursinha [07:07] Good morning [07:07] good morning ! [08:44] ali1234 no [08:44] ali1234: but preliminary stuff is ready [08:45] ali1234: grab the xorg-server from that ppa and you only need to compile your own ddx and own mesa.git [08:45] good morning desktopers [08:45] ohai [08:45] hey seb128, morning mlankhorst [08:46] haithere [08:46] pitti, hey, wie gehts? [08:47] seb128: not too bad; a bit tired still from the weekend [08:47] it was a long car drive there (Fri) and back (yesterday) and a quite long/intense Saturday [08:47] there == Hamburg, we visited a friend there who recently moved there [08:47] oh, what did you do? [08:47] I see [08:48] so we did a sightseeing tour, a port cruise, and the Reeperbahn in the evening [08:48] hey seb128, pitti! [08:48] bonjour didrocks ! as-tu eu un bon week-end ? [08:49] pitti: bon week-end! we bought some new furnitures for the kitchen, so no visiting for us :) [08:50] also, got unity broken over the week-end, trying to trick popey to see if he can reproduce [08:50] didrocks, lut [08:50] oh, "fun" [08:51] * seb128 notes to not upgrade [08:51] seb128: I guess you're fine, it was published the 7 of January [08:51] seb128: I had unity marked on hold [08:51] oh, ok [08:51] I unmarked it on Friday, dist-upgraded [08:51] yeah, no issue here then [08:51] but didn't reboot [08:51] until Saturday :p [08:51] what's the issue? [08:52] the stacktrace is weird: http://paste.ubuntu.com/6784761/ [08:52] I don't see anything stricking/hurting my eyes here [08:52] also, I tried to report it with apport-bug [08:52] but I saw the apport-gtk process running [08:52] but no windows [08:52] not sure how since we have this whoopsie thing how it works [08:53] and 30 minutes after trying to report it, nothing on errors.ubuntu.com [08:54] so, I reverted to previous nux and unity (the one not rebuilt against glew 1.10) and it's working [08:58] glew issues, fun [08:59] yeah, I see nothing showing that on the stack [08:59] * didrocks is sad the once again, foundation did the upload totally untested with us and uncoordinated [08:59] that* [09:00] seb128: I'm still wondering why apport-gtk didn't show up though, even if whoopsie is enabled, when I tell "report it", it should use apport (as enabled on my system), right? [09:00] * didrocks is totally lost with whoopsie [09:00] but I see a .upload and .uploaded and I have no idea where this is :/ [09:01] didrocks: where do you tell it "report it"? that should already be apport [09:01] didrocks: system settings -> privacy -> diagnosis -> show previous reports [09:01] didrocks, not sure it's "totally untested", that update seems to be working for most of us [09:01] didrocks: that should lead you to "your" errors.u.c. page [09:01] seb128: well, I doubt it ran under the 3 otto machines [09:02] seb128: hence the "with us" [09:02] pitti: ah nice, let me see :) [09:02] yeah, not likely, but to be fair I don't test every archive upload I do on the otto machines either [09:02] seb128: yeah, but you don't upload nux and unity :) [09:02] didrocks: that web page is rather unfriendly ATM, it only shows hashes; it should show projects, dates, etc. [09:03] seb128: if I want to transition something on glib, I would ask you first ;) [09:03] pitti: hum, let me click around [09:03] yeah [09:03] didrocks, in any case, e.u.c doesn't rank any unity bug [09:03] seb128: compiz-core crash? [09:04] seems it's https://errors.ubuntu.com/oops/a8ea9f18-8068-11e3-a0f6-2c768aafd08c [09:04] pitti: but it's not retraced from what I see [09:04] yeah, or retrace failed [09:04] pitti: and there is no more apport asking us if we want to file a bug in launchpad, if apport is enabled? [09:05] I noticed that on a lot of whoopsie reports, it seems to have trouble with trusty [09:05] morning! [09:05] didrocks: not ATM; we might actually re-enable it for alpha-2; seb128, WDYT? [09:05] didrocks: it's still disabled from the saucy release [09:05] we traditionally re-enable LP crashes around alpha-2 [09:06] Laney, hey, how are you? [09:06] pitti: it's a flag on the desktop which isn't /etc/default/apport? [09:06] hey Laney [09:06] morning Laney [09:06] didrocks: no, it's problem_types in /etc/apport/crashdb.conf [09:06] didrocks: if you comment out/drop that line, it'll report crashes to LP [09:06] pitti: so, don't set any default? [09:07] pitti, re-enabling, wfm, I've no strong opinion, I look more at whoopsie than launchpad nowadays [09:07] not enough free slot to keep with launchpad [09:07] ^ yes, that's why I don't want to re-enable LP crashes too early [09:07] pitti: thanks, I was really wondering wth with apport enabled, I couldn't get a "normal" apport report in launchpad! ;) [09:08] didrocks: we don't change /e/d/apport any more since we use whoopsie [09:08] ok, so apport always "report" [09:08] as evan has a rather strong opinion about leaving this enabled in stables [09:08] and this file change the direction [09:08] right [09:08] destination* [09:08] either to errors.u.c. only, or additionally to LP [09:08] ok, making sense, at least, that would have teach me a little bit how this works now :) [09:08] thanks pitti! [09:09] now on why it's not retraced [09:09] * didrocks is pinging ev [09:10] seb128: I hope it's really only my configuration, will get a confirmation with popey (he's upgrading his x220) [09:10] hey seb128 pitti didrocks! [09:10] I tried to drop all my favorites under /com/canonical in case it was a launcher icon for instance [09:10] sounds like a fun monday start ;-) [09:10] (as it seems to fails here) [09:10] didrocks, well, at least it's not most of us, if it's there for some 10 days (nobody complained) [09:10] Laney: well, it's fun on the phone ML as well ;)) [09:11] didrocks, did you try a guest session? [09:11] seb128: that's what I thought. I hope though it's not this graphic-card related [09:11] seb128: hum… I lost my old habits I guess [09:11] * didrocks should try [09:14] seb128: ok, now I'm puzzled [09:19] hum, ok, it restarted here as well now [09:19] both guest and my session wth [09:19] * didrocks unlogs and comes back [09:22] ok, can't reproduce it anymore [09:22] restarted lightdm [09:22] tried 2 monitors, then one, like yesterday [09:23] didrocks, weird, maybe it was something in your launcher config? [09:23] so not related to my graphic card as I thought :/ [09:23] seb128: no, I tried to empty it [09:23] dconf dump /com/canonical (even indicators then) [09:23] then dconf reset -f… [09:23] was still crashing [09:23] then, I reloaded my config [09:23] and was again crashing of course [09:24] weird [09:25] I'm out of ideas and don't like to call that a "non issue anymore" [09:25] this was just a ploy to get me to update to trusty wasn't it? :D [09:25] popey: yeah, all that in that end! :) [09:25] popey: sorry, I was really thinking it was x200-configuration related ;) [09:25] x220* [09:25] np [09:26] that's a strange one in any case [09:26] ok, I'm dropping offline for like 10min [09:26] brb [09:38] ooo [09:40] mlankhorst: do you know about this kind of thing http://paste.ubuntu.com/6784936/ ? [09:51] yeah something messed up, no idea what yet :P [09:51] Laney: probably some kernel upgrade? [09:52] there was one [09:53] apport is offering me a bug report [09:56] not sure a bug report would be helpful [10:05] Sweetshark, happy birthday! [10:12] birthday bumps [10:22] seb128: thx. btw I put 4.2.0~rc2 in the PPA for trusty (and the always awesome ricotz already bumped it back for saucy too). Its looking good in general and is also building on trusty/armhf. There was one hickup with a dependency of a transitional, but that is fixed now. [10:22] Sweetshark, \o/ [10:23] seb128: so I guess, what I am trying to say is: If nobody is screaming protests, IMHO we should go to trusty/main with that one as it is now. [10:25] do pitchforks count? [10:25] seb128: I will walk through the remaining --with-system-$foo switches and see if it would be good and worth the risk to debian-sync/MIR/whatever them and build against those, but we shouldnt do all that with one upload. [10:26] Sweetshark, +1 from me to upload to trusty [10:26] mlankhorst: Dont attack LibreOffice with forks. We'll beat you with our experience on that one. [10:26] Sweetshark, btw we should maybe take the new poppler (soname transition) before uploading libreoffice ... you said libreoffice didn't seem to be using the API that change, to you want to test build that or something? === vrruiz_ is now known as rvr [10:27] seb128: testbuilding sounds good. When is the poppler transition estimated to take place? [10:27] Sweetshark, "whenever it's convenient for libreoffice" [10:28] Sweetshark, the other rdepends are easy to rebuild, it's only 1 function that changed [10:29] seb128: is that new poppler upload in some ppa or something already? or already in -proposed? [10:30] Sweetshark, no, the diffs are on https://bugs.launchpad.net/ubuntu/+source/poppler/+bug/1256627, I can upload to the desktop team ppa if that helps though [10:30] Launchpad bug 1256627 in poppler (Ubuntu) "New upstream release, merge with Debian" [Wishlist,New] [10:33] seb128: the other thing I tried to solve is to get back my ~90minutes binary builds locally on the notebook. sadly that didnt work out. I optimized my pbuilder setup, disabled xz compression of .debs (which take another hour), but somehow could get ccache to do anything else than cache misses. So still at 7,5 hours builds, which is a pain. :( [10:33] seb128: diffs should do, thx. [10:33] Sweetshark, great, let me know how that goes ;-) [10:34] seb128: willdo. ... In ~14 hours ... === MacSlow is now known as MacSlow|lunch [12:25] davidcalle: Hey dude, scopes... if I type in weather on the phone or desktop (13.10 and 14.04) in theory is should show me the location for my area correct? If so I think it might be failing :( [12:27] seb128: that poppler patch does a "dh_install --list-missing\n dh_install: libpoppler44 missing files (usr/lib/*/libpoppler.so.44*), aborting" here ... [12:27] Sweetshark, what version did you build? [12:28] Sweetshark, http://poppler.freedesktop.org/poppler-0.24.5.tar.xz ? [12:28] seb128: apt-get source poppler && patch -p1 patch-from-1256627 [12:28] seb128: argh, k [12:28] Sweetshark, right, that patch is a new version update [12:29] that's a bit confusing [12:29] basically take that tarball and copy the debian dir over [12:30] seb128: yeah, I just saw the huge patch and thought its a full diff (not just ./debian) [12:31] davmor2, hey. There has been some unexpected API changes from the provider, it's being fixed. [12:32] davidcalle: ah nice okay if it is known I shall go back to my corner :) [12:35] seb128: are you sure about the poppler-0.24.5.tar.xz? because the patch talks about 24.3->24.4 not 5. [12:37] Sweetshark, .5 seems a small bug fix over .4, the API change is in .4 [12:37] Sweetshark, so either would do [12:38] davmor2, hehe, thanks for the feeback, though ;) (fun fact about this bug : it's more likely to happen if it's past mid-day in you local timezone, can you confirm it is ?) [12:39] davidcalle: 12:38 I will check tomorrow first thing and see if works then though to confirm [12:39] davmor2, thanks :) [12:52] seb128: twas because the patch didnt apply cleanly on the changelog, so I was left with the old version there m( === Sweetsha1k is now known as Sweetshark === MacSlow|lunch is now known as MacSlow [12:57] * Laney yays at the glib-networking flaky tests getting fixed [13:42] * Sweetshark finds missing dependencies in the upstream build. [13:43] With 32 threads you find all those that others miss. [13:43] * Sweetshark wants your sympathy. [13:57] hmmm, without l10n, the build is 13GB -- I could squeeze that in a tmpfs ... [14:18] sil2100: hi, do you know if your grilo patch was forwarded? [14:19] * mlankhorst compiles Sweetshark into plain english [14:19] http://www.plainenglish.co.uk/ ? === jpds_ is now known as jpds [14:30] Laney, btw did you see the backlog from the sdk guys on friday evening? they replied to your autopilot error question [14:30] seb128: yes, I replied [14:30] ok, I don't IRC proxy so I didn't see that [14:30] I'm waiting for elowotsit to come online [14:30] ok [14:30] basically I don't think it's helpful as this is not a listview [14:30] the licenses panel? [14:31] about [14:31] ah, I though you said you had issues with the licences list [14:31] It's Column { ListItem ListItem ListItem ... } [14:31] it can't select the button to get into the page [14:31] the license page [14:31] I see [14:32] I wonder if autopilot is the right tech for those tests [14:32] it's a tech for integration tests [14:32] we should maybe look at doing the UI tests with the qt runner stuff [14:33] qmltestrunner (was looking for the name) [14:34] dunno, some integration testing is definitely valuable [14:35] right [14:35] we should have both imho [14:36] the tests checking that e.g panels load by clicking and checking there is a title are purely UI ones [14:36] they probably don't need autopilot [14:38] yeah I agree we should have both [14:38] the uitk is a good example of QtQuickTest unit tests [14:39] there was an email about that on the phablet list back then [14:39] the thinking was mainly "if you just need to test UI/app logic, use the qt tools, it's less work and less slowness/delays", "use autopilot for integration tests though" [14:42] oops, forgot to have lunch, brb! [14:42] haha [14:42] Laney, enjoy ;-) [15:13] seb128: dpkg-buildpackage: full upload (original source is included)\n real58m37.657s [15:13] seb128: seems that libreoffice does not barf all over poppler 0.24.4 [15:14] great! [15:49] Laney: our poor tests :( [15:49] hah [15:50] danw fixed the glib-networking one [15:50] we need a revert-testcase policy [15:50] actually, probably we don't need that [15:51] since it's quite possible that new tests or changes in existing tests are finding real bugs [15:52] * desrt is just getting sick of the 90%+ experience of "it's just the test doing dumb stuff" [15:52] I'd like to know what's so pathological about Ubuntu's buildds and the CI environment [15:52] Laney: heavy load, i think [15:52] when i 'make distcheck' it's the only thing i'm doing [15:52] the buildds only build one thing at a time [15:52] i thought they were vms [15:52] non-virtually [15:52] huh [15:52] PPAs are virtualised [15:53] but not the main distro builders [15:53] maybe they're just old crappy machines [15:54] maybe for weird arches [15:54] but i guess our x86/amd64 builders are pretty decent [15:56] hey [15:57] didn't we have a thing a while ago where we ran some gnome upstream testsuites? [15:57] * ogra_ is pretty sure you wont find much faster armhf HW than our buildds :P === alan_g is now known as alan_g|tea [16:00] pitti: ^ what happened to these jhbuild runs? [16:00] did they go away with the advent of continuous? [16:02] Laney: mostly yes; and we didn't really keep up the manpower for maintaining those with the changed focus on uphone [16:03] nod === alan_g|tea is now known as alan_g [16:20] i love you posix [16:20] EINVAL The value specified by cond, mutex, or abstime is invalid. [16:21] what's wrong? who knows.. could be anything! [16:30] Hmm, I've got evolution-data-server-uoa installed, but my google account still does not appear in evolution :( [16:31] with ubutnu 14.04 [16:31] xclaesse, did you just add it? [16:31] seb128, I realized that I had e-d-s-goa installed and removing it made -uoa replace it [16:32] seb128, I tried deleting and recreate my google account, but still it does not appear in evo [16:32] hum [16:32] can you reinstall -goa? [16:32] (e.g have both) [16:32] then try again? [16:32] I wonder if that issue is back (needs to look that split again) [16:33] it might be that the goa binary has stuff that should be common to both (we had the issue in the past and went forth and back, so it might be buggy again in trusty) [16:33] shrug [16:34] Laney 0 - 1 webkitgtk [16:34] :-( [16:34] haha [16:34] it's way more than 1 [16:34] seb128, success ! [16:35] seb128, yep, installing -goa made it work [16:35] xclaesse, ok, thanks for reporting the issue [16:35] or maybe it is because the account is configured into goa? dunno... [16:36] seb128, is there a trick to get the GOA panel on ubuntu ? [16:36] xclaesse, I need to check, but iirc it was because the goa binary has code needed by uoa [16:37] xclaesse, XDG_CURRENT_DESKTOP=GNOME gnome-control-center [16:37] I see that goa-daemon is actually running [16:37] yeah, if it's installed it's running [16:37] some stuff dbus activate it iirc (it's not installed by default though) [16:38] ok, removed all my accounts there, and evo still has the google account, so it's from UOA :) [16:38] good ;-) [17:07] https://bugs.webkit.org/show_bug.cgi?id=126985 [17:07] bugs.webkit.org bug 126985 in Web Template Framework "Source/WTF/wtf/Atomics.h:300: Error: bad register name `%bpl'" [Normal,Resolved: fixed] [17:07] I might wait to see if it builds everywhere else though [17:15] hm, 14.04 is not ready with CSD :( [17:15] deja-dup-preference does not have decorator [17:16] and close button is at right, without ubuntu style [17:18] seb128, hm, when deja-dup is doing its automatic backup, there is no icon in the ubuntu launchers anymore [17:18] no indication that something happens [17:18] that's a regression since 13.10 [17:19] mterry, ^ [17:19] curious... [17:19] I will look [17:19] there were a nice progress bar on the launcher icon previously [17:19] mterry, btw, what are you doing working on a national U.S holiday? [17:19] maybe it removed something, because I don't find the icon it all apps [17:20] and it's not in the gnome-control-settings neither [17:20] seb128, :) eh, I'm not working hard [17:20] xclaesse, mterry: seems to be an issue with the unity-control-center transition [17:20] I've the panel in unity-control-center, not in g-c-c.real [17:21] xclaesse, we started a transition but it's not done yet, seems to be a side effect [17:21] seb128, ah, unity-control-center is not installed [17:21] xclaesse, renaming our patch g-c-c to u-c-c so the GNOME remix can provide a less patched/more uptodate version [17:21] I've got gnome-control-center-unity [17:21] xclaesse, yeah, it's not supposed to be yet, some panels didn't transition ... well you can install both [17:22] seb128, ok, backup is in unity-control-center :) [17:22] yeah [17:23] that doesn't explain the lack of launcher integration while it's backuping though [17:26] Huh, "Backup" should show up in both right now [17:28] tbf, maybe my system isn't correctly upgraded, update-manager just crashed when I tried to upgrade from 13.10 [17:28] so I did a plain apt-get dist-upgrade [17:30] Hmm, 2014, and still the firefox window asking for password when accessing an https page is totally hidden [17:30] have to move around all my windows to see where it's hiding behind [17:30] is it that complicated to have everything in alt-tab ? [17:31] something with transiant/modal windows [17:43] have a good even everyone === alan_g is now known as alan_g|EOD [19:07] Hi folks, with the last update from gnome3-team (screensaver) is no more working [19:07] some one have the same issue? [19:08] also change/switch user is no more working [19:19] looks like gnome-screensaver doesn't run with the session [19:41] Hm, evolution often silently closes because it crash and no crash report seems to be created, dunno why... [19:41] here is the backtrace though: http://paste.ubuntu.com/6787703/ [19:41] ubuntu 14.04 [19:42] maybe I told apport to stop reporting evolution issues a while ago, don't remember [21:32] xclaesse: I see similar results in rh's bugzilla - I'd report it directly upstream if I were you [21:33] we don't really have code patches to evo === m_conley_away is now known as m_conley === m_conley is now known as m_conley_away