=== thumper-gym is now known as thumper-hacking === thumper-hacking is now known as thumper [04:11] Good morning [06:54] ricotz: I want to run Ubuntu on my Android board. I have IFC6410 development board [06:55] ricotz: I have used the kernel in the Android source and created the ubuntu *.deb packages [06:56] ricotz: Can you please help me? [06:58] abhishek_, sorry, i don't have an armhf setup here [06:59] ricotz: Can you please give me some guildlines [07:01] abhishek_, obviously you already got some hints here https://lists.launchpad.net/ubuntu-phone/msg04938.html [07:02] abhishek_, there is also an #ubuntu-touch channel for those specifics [07:03] ricotz: I don't want to run ubuntu touch....I want to run Ubuntu desktop on this board [07:03] ricotz: The ubuntu as we usually see on our desktop or laptop machines [07:04] i see [07:05] and the stock kernel isn't working for you? [07:06] ricotz: I have extracted the boot.img image of the Android ...and found that it has zImage (kernel ) and ramdisk ....ramdisk contains init scripts which actually initiallizes the board [07:06] ricotz: I must be having some similar kind of ramdisk to initializes my board .....where to create or get this ramdisk [07:07] ricotz: I have created the zImage ...and linux-image.deb and linux-headers.deb packages [07:08] ricotz: I have extracted linux-image.deb package and saw that it contains all the *.ko files [07:08] abhishek_, as said, i don't have such a setup here, so i am not that experienced in that regard [07:09] ricotz: I must pack the zImage with the ramdisk to create the boot.img image which I can load in the boot sector of the device .......my problem is how to create the ramdisk [07:10] ricotz: ok ...from where should I get the help ??? [07:10] abhishek_, if your board is quite popular there is for sure some guide somewhere, maybe for debian which might be similar [07:11] ricotz: The board is new ...I think I am the first one which is doing this for this board [08:55] Moin! [08:59] seb128: I changed the shortcut to Ctrl-space in ibus, and uploaded to trusty for now. so in unity/gnome super-space works, and in kde ctrl-space works. Do you think this is acceptable? [09:03] happyaron, hey, hum ... didn't you say the other day that super-space wasn't working for you, even after changing ibus? [09:04] it works for me, while I need to clean up configs in my home dir [09:04] rm -rf ~.*; and re-login [09:04] sorry, rm -rf ~/.* [09:05] morning [09:05] weird, it worked for me without that [09:05] happyaron, in any the change makes sense, so +1 [09:05] Laney, hey [09:05] good morning desktopers ;-) [09:06] seb128: so push the same change as SRU to saucy? [09:06] yes, I guess it makes sense [09:06] ok, will do that [09:07] seb128: and it's wired that the ibus-ui-gtk3 crash does not appear in unity anymore after installing all the SRUs, but appear again after I installed kubuntu-desktop, clear configs, and login to KDE. [09:08] so maybe it's just an issue on clear config? [09:08] did you figure out if that one is the same that I assigned to you on launchpad? [09:09] seems yes, but I still don't know the cause. [09:09] after clearing configs, login to unity does not have such crash [09:09] but clear configs then login to KDE will trigger the crash. [09:10] it happens under unity sometimes [09:10] I had it in guest sessions [09:11] ok [09:11] will dig more into it. [09:11] though not recently [09:11] maybe gsd does some config init that workaround it or something [09:11] did you try to just rm the config and restart ibus manually in a session? [09:12] trying [09:12] no crash in unity [09:17] and no crash in KDE, :-/ [09:34] happyaron, was it ibus-ui-gtk that was hitting the bug for you? [09:34] yes [09:40] happyaron: I want to create the initramfs for the ARM board from my x86 machine ? ... [09:40] happyaron: Can you please help me [09:41] abhishek_: I'm not familiar with doing that... [09:41] happyaron, valgrind lists that issue, not sure if that can lead to the bug [09:41] ==7168== Conditional jump or move depends on uninitialised value(s) [09:41] ==7168== at 0x4381A97: gtk_tray_icon_get_colors_property (gtktrayicon-x11.c:560) [09:41] ==7168== by 0x438233A: gtk_tray_icon_update_manager_window (gtktrayicon-x11.c:859) [09:41] ==7168== by 0x438290F: gtk_tray_icon_constructed (gtktrayicon-x11.c:266) [09:42] happyaron: ok ....I am not getting help ....can you please let me know who can help me? [09:44] abhishek_: sorry, dunno... [09:44] seb128: I'll try later, going for dinner now. [09:45] happyaron, enjoy! [09:46] mvo, hey, do you have any idea about https://bugs.launchpad.net/ubuntu/+source/update-manager/+bug/1024590 ? [09:46] Launchpad bug 1024590 in update-manager (Ubuntu Saucy) "update-manager crashed with AttributeError in _on_download_changed(): 'NoneType' object has no attribute 'get_value'" [Medium,Confirmed] === tkamppeter_ is now known as tkamppeter [11:36] seb128: saw the verification-failed on that presentation on 3.5.7-0ubuntu5, commented on it -- verification of a perfomance issue is tricky business, it seems: how much faster does it have to be to be verified .... === vrruiz_ is now known as rvr [11:52] Laney: are we running the glib tests installed? I'm thinking about switching the gtk tests over, because they're causing trouble when running in the build dir [11:52] I'm wondering how that would work in debian-land [11:52] both at build and install time [11:52] oh, the same tests? [11:52] madness [11:52] well, make check and then installed tests [11:52] but it works out to be the same thing [11:53] so you know the thing you just built works, and that it keeps working [11:53] right, but this won't fix my problem for me then :) [11:53] I guess I'll just fix the non-installed case and switch later [11:53] yeah, make check is supposed to keep on working :P [11:54] in practice people have been getting one or the other wrong [11:54] right [11:54] I was just looking for the easy way out [11:56] is it an actual test failure or some problem with adding the installed test support? [11:58] it's a failure because it accesses installed settings schemas [11:58] it fails when you don't have the new keys [11:58] or no gtk installed === MacSlow is now known as MacSlow|lunch [12:01] * Laney nods [12:16] * walters thinks it's worth the pain =) [12:18] walters: installed tests, you mean? [12:18] yeah [12:18] installed tests work in this case. It's in-tree make check that fails [12:18] otherwise I totally agree of course ;) [12:19] this case is in-tree but srcdir != builddir? [12:21] yes, but that shouldn't matter. It's never accessing the right schemas [12:21] unless they're installed [12:22] ah...right [12:22] basically that never gets tested due to how jhbuild works [12:22] yep [12:22] I'll come up with a test as soon as I got this package updated [12:23] * larsu has been yak-shaving enough this week [12:23] s/test/patch [12:24] on the bright side: the new default-values tests found a bug in one of the ubuntu patches [12:45] Laney, could you have a look to https://bugs.launchpad.net/ubuntu/+source/gnome-settings-daemon/+bug/1065979? it's in the sponsoring queue and your sponsored the previous version of the change [12:45] Launchpad bug 1065979 in gnome-settings-daemon (Ubuntu) "external/internal monitors mirrored on boot when laptop lid is closed" [Wishlist,Confirmed] [12:46] seb128: ok, could you assign me? [12:46] mlankhorst, hey, could you look at https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/1197921 ? [12:46] Launchpad bug 1197921 in pixman (Ubuntu) "LibreOffice spreadsheet causes full Xorg crash with Anti-Aliasing enabled" [High,Triaged] [12:46] sponsoring ninja [12:46] Laney, sure, should I keep the sponsors as well or unsubscribe those? [12:47] keep it, someone else might do it meanwhile [12:47] Laney, down to under 70 today, still not great but that's already better ;-) [12:48] some of the red ones have been there for ages [12:50] blame foundations [12:50] they are mostly stuff they know about, I don't feel competent to sponsor those [12:51] e.g debian-installer, sysvinit, partman, acpi-support [12:51] yeah I tried pinging a couple of times [12:51] efitbootmgr [12:54] * seb128 tries some #ubuntu-devel nudging [13:03] jdstrand, mdeslaur: I guess you guys know but our apparmor is outdated (2.8.0 where upstream is at 2.8.2) ... do you guys plan to update? ;-) (/me trying to clean outdated packages from versions) [13:05] seb128 (and mdeslaur): heh, yes we know. we are actually *far* ahead of 2.8.2. plan is to get back in sync with trunk (2.9) and drop 70 patches or so :) [13:06] that isn't going to happen today though [13:06] jdstrand, is that happening this cycle? (or should I just mark 2.8.0 as "uptodate" in our status page to get if off the list) [13:07] it is happening this cycle [13:07] great [13:07] jdstrand, thanks [13:08] seb128: it will 100% happen by December (when we land our IPC and LXC work). but for now, it is actually up to date. so however you want to handle it === alan_g is now known as alan_g|lunch [13:09] jdstrand, I'm probably going to just add a note about it, thanks [13:10] sounds fine [13:17] chrisccoulson, hey [13:18] chrisccoulson, https://bugzilla.mozilla.org/show_bug.cgi?id=924395 ... can you tell me in what version of firefox that's going to land? ritz added a sponsoring request to SRU this fix, do you think that's worth a SRU or that we should just wait the next security update/firefox update? [13:18] Mozilla bug 924395 in General "Gnome proxy settings, address conversion to IPv6 for "ignore hosts" doesn't convert netmask" [Normal,Resolved: fixed] [13:19] chrisccoulson, that's https://bugs.launchpad.net/ubuntu/+source/firefox/+bug/1226537 [13:19] Launchpad bug 1226537 in firefox (Ubuntu) "System proxy settings doesn't work in Firefox when URL is a numeric IP address" [High,Confirmed] [13:28] cyphermox, do you want to have a look to https://bugs.launchpad.net/ubuntu/+source/ejabberd/+bug/1097366 ? You commented by then saying you were interested but needed a config to test, there was one provided since [13:28] Launchpad bug 1097366 in ejabberd (Ubuntu Raring) "Pidgin client digest-uri checks fail with multiple virtual hosts" [Medium,Confirmed] === alan_g|lunch is now known as alan_g [13:51] seb128, can you unsubscribe ubuntu-sponsors from bug 1247877? I'll submit it directly to Debian [13:51] Launchpad bug 1247877 in python-dateutil (Ubuntu) "Enable autopkgtest" [Medium,Triaged] https://launchpad.net/bugs/1247877 [13:52] jibel, done, thank you ;-) [13:54] yw === MacSlow|lunch is now known as MacSlow [14:02] seb128: do you want to postpond upgrading to ibus 1.5.4, for CVE-2013-4509 [14:03] I have some clue to the crash now, but the code is written in vala and is a bit difficult for me to debug. It's very probably introduced in ibus-541492-xkb.patch, which was cherry-picked from Fedora 19. [14:04] happyaron, @CVE: do you have a reference? google returned https://bugzilla.redhat.com/show_bug.cgi?id=1027028 which states it might be an issue since 1.5.2 [14:04] bugzilla.redhat.com bug 1027028 in vulnerability "CVE-2013-4509 ibus: visible password entry flaw" [Low,New] [14:04] seb128: see https://groups.google.com/forum/#!topic/ibus-user/mvCHDO1BJUw [14:05] http://people.canonical.com/~ubuntu-security/cve/2013/CVE-2013-4509.html [14:05] seb128: only 1.5.3 isn't affected [14:05] seb128: ^ [14:05] jdstrand, happyaron: thanks [14:05] happyaron, let's wait for the fix to update then [14:05] ok [14:06] so we don't want to merge 1.5.4-1. [14:06] yup [14:06] jdstrand, yeah, that's what happyaron was asking about ... well we want to merge but we should wait for the next version or to have a patch for the CVE, before uploading [14:06] yes [14:06] happyaron, @vala: if you need help we have people on the channel who know it pretty well, so feel free to ask [14:08] thanks, I'm still new to it, will ask later if I can't make it. [14:30] timchen119, hey, are you there? [14:43] seb128, hi :) [14:44] ricotz, hey, how are you? [14:44] seb128, was it intended to drop the libunity support of geary? [14:44] pitti, hello! You know autopkg, eh? :) Do you know what this failure in deja-dup is saying is the problem?: http://10.189.74.2:8080/job/trusty-adt-deja-dup/13/ARCH=amd64,label=adt/ [14:44] ricotz, no, how did that happen? missing build-depends? [14:44] seb128, i am good :), thanks, i hope the release stress has settled [14:45] seb128, i guess so [14:45] ricotz, yeah, it's better, another busy cycle started though [14:46] libmessaging-menu-dev was added back but not libunity-dev [14:46] ricotz, thanks for pointing it, is there a bug report about it? [14:46] seb128, i am still counting for gtk 3.10 ;) [14:47] seb128, no bug report, i just stumbled over it [14:47] ricotz, you might want to help that to happen and help looking at theme issues etc? [14:48] seb128, larsu promised to port the important menu patch which is crucial to have [14:48] ricotz, he did [14:48] seb128, besides that i guess there are only theming issues [14:48] ricotz, right, which is where we would welcome help [14:49] ricotz: ya I'm done [14:49] seb128, i am hoping porting it wont break abi/api so it can be easily used on saucy [14:49] I'm still fighting with some build issues (testing mostly) [14:49] larsu, nice! [14:49] i hope it is compatible with ido in saucy [14:49] of course [14:49] good :) [14:51] larsu, what issues? [14:51] i ran into trouble having those symbols exported and forced it the hard way [14:51] in-tree testing is broken for some tests [14:52] and also uncovered a bug in one of our patches :) [14:52] for some reason the export macro didnt work [14:52] I'm disabling the offending tests [14:52] i see [14:52] I'll reenable them later. I just want to get the package building asap [14:53] which symbols? The ubuntu_* ones? [14:53] yes the interface symbols [14:53] what was the problem? It works fine here [14:54] ah, you need to tag functions for exporting now with GDK_AVAILABLE_IN_* [14:54] larsu, no idea why it didnt work https://launchpadlibrarian.net/155575870/gtk%2B3.0_3.10.2-0ubuntu1~saucy3_3.10.2-0ubuntu1~saucy4.diff.gz [14:54] larsu, i know [14:54] heh, yeah it took me some time to figure that one out: config.h was not included [14:55] don't worry about it though, I'll have the fully updated patch up in a bit [14:55] ah :\ [14:55] if the tests pass this time around [14:55] you actually want to include gdk.h [14:55] that alone didn't do it [14:55] and it already is through gtkwidget.h [14:55] alright [14:56] larsu, feel free to point me to the package later :) [14:56] will do [14:57] GDK_AVAILABLE_IN_3_10 expands to "extern" [14:57] oops, ignore me [15:04] larsu: are you adding an installed-tests package? [15:05] Laney: not yet [15:05] nod [15:05] I'm simply trying to get the package building as is [15:05] but I'm running into more and more issues :-/ [15:06] at least the yak will look good when you're done :( [15:06] haha [15:08] when was org.gtk.MountOperationHandler added? [15:09] larsu, not recently, why? [15:09] it's the cause of some test failures. GtkMountOperation tries to access that name on the bus and crashes if it's not there [15:09] larsu, https://git.gnome.org/browse/gtk+/commit/?id=44fd03eb47b9492ffaf32ae6dfcca00b7d5a7618 [15:13] well, this would have bitten us later. I'm glad the tests caught it [15:13] not glad that I have to write a patch for it _now_ === mjohnson15 is now known as MalcontentMatt === gatox is now known as gatox_lunch [15:30] seb128: my bad, this is because we don't have access to the bus at all in fakeroot [15:30] and those tests don't run under gtestdbus or similar [15:31] larsu, welcome to the fun of the packaging world :/ [15:31] haha thanks [15:31] larsu, I guess you need dbus-test-runner or similar? [15:31] yes [15:32] these tests normally just use the session's dbus and dconf [15:32] scary. === alan_g is now known as alan_g|afk [15:37] larsu, but you get testing of dbus for free this way ;) [15:38] ogra_: unless there's no bus around, which causes all tests to fail [15:39] details :) [15:39] haha [15:41] what does dbus-test-runner do over dbus-launch? [15:43] I don't know, I'm using GTestBus [15:44] so I've added a patch, how can I make debuild pick it up without rebuilding everything? [15:45] Laney: Hi, time for a follow-up remark about the u-s-s discussion last night? [15:46] larsu: I usually work with the upstream build system until happy, then switch to the packaging [15:46] maybe 'debuild binary' though? I think seb128 works more in that way [15:46] GunnarHj: maaaaybe [15:46] working on AS atm btw (a different patch, but getting warmed up) ;-) [15:47] Laney: Let's try, then. ;-) [15:47] If the final decision is to use 'the simplistic approach', i.e. just set a locale, there is (at least) two options. You can call both SetLanguage and SetFormatsLocale. But those methods in Ubuntu are patched with separating display language and formats in mind. An alternative way is to not make use of accountsservice's Language property at all, and write to ~/.pam_environment directly. In that case it would be sufficient t [15:47] o only set LANG. [15:47] Laney: that's what I did. The upstream build system is happy but the package build failed on the tests [15:47] oh well, I guess I'll just recompile [15:48] try dpkg-buildpackage -nc [15:48] GunnarHj: What if things want to read it out of AS? [15:48] Like the greeter or whatever [15:49] Laney, http://gould.cx/ted/blog/Reimagining_dbus-test-runner [15:50] Laney, dbus-launch relies on having a dbus configuration on the system afaik, also the test runner has some flexibility in the options/behaviour ... tedg would sell it better probably though ;-) [15:51] I bet it is more flexible somehow, probably my bad for not using it [15:51] not suitable for upstreaming to debian though :P [15:51] ? [15:52] BTW, libdbustest has dbus-mock support! Way cool. [15:52] Laney: Then it's best to use the a-s methods, of course. But please note that with the Ubuntu patches, the Language property stores a language code (ll_CC), not a locale name. [15:52] GunnarHj: Yeah [15:53] Laney: That's handled on the desktop, and if you e.g. use the same lightdm version on the phone as on the desktop, it should be ok. [15:53] That's the eventual idea [15:53] we should design solutions with that in mind [15:54] Laney: Ok. [15:54] The current code in u-s-s does call both of those methods [15:54] attente wrote it, but I think it passes them the right things [15:55] Laney: Actually, SetLanguage expects a language code, but is able to deal with a locale name also. ;-) [15:56] Laney, Huh, so it seems that dbus-test-runner never made it into Debian. It got to ITP, but not further. There was a belief that it required Ubuntu changes to DBus, but that's not correct. [15:56] Laney, http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=677271 [15:56] Debian bug 677271 in wnpp "RFP: dbus-test-runner -- Runs tests under a new DBus" [Wishlist,Open] [15:57] tedg: Maybe you could post there (nnnn@bugs.debian.org) to correct the record [15:58] Laney: Given that e.g. lightdm queries a-s for language, It sounds as if all is well then. You can disregard the idea above. [15:59] GunnarHj: Alright then, thanks for your thoughts [15:59] Laney: yw [16:00] bregma, the bug you just reassigned to ido is fixed in https://launchpad.net/ubuntu/+source/ido/13.10.0+14.04.20131106-0ubuntu1 [16:01] bregma, it was in trusty for half a day, but I guess it's enough for users to grab the update and still run it === gatox_lunch is now known as gatox [16:02] just so long as it doesn't get lost [16:03] bregma, well, I'm just pointing so you don't spend time trying to debug to figure out that it was already handled === alan_g|afk is now known as alan_g [16:45] hum [16:45] pitti, did you see " GVFS-RemoteVolumeMonitor-WARNING **: remote volume monitor with dbus name org.gtk.Private.UDisks2VolumeMonitor is not supported" errors before? [16:45] (looking to https://bugs.launchpad.net/ubuntu/+source/nautilus/+bug/1241811 which seems to bites some users) [16:45] Launchpad bug 1241811 in nautilus (Ubuntu) "USB disk not automatically mounted after upgrade to 13.10" [Low,Incomplete] [17:03] seb128: well I already did my full analysis there :P [17:03] was hoping to get feedback from upstream [17:42] seb128: hi! do you have the ability to move around stuff in the client track for vUDS? [17:43] jono: ^ [17:43] jdstrand, tied up right now, mhall119 can help [17:48] jdstrand, I can yes [17:48] seb128: cool, can you switch these two sessions: http://summit.ubuntu.com/uds-1311/meeting/22061/client-1311-user-data-encryption/ on 11-20 and http://summit.ubuntu.com/uds-1311/meeting/22083/client-s-system-settings-panels/ on 11-19? [17:49] jdstrand, do you need them swapped or do you just want the settings slot for your data session? [17:50] the latter. right now user data encryption conflicts with the upstart roadmap [17:50] jdstrand, ok, done [17:51] seb128: thanks! :) [17:51] yw! [17:51] mhall119, jono: fyi, seb128 helped me === alan_g is now known as alan_g|EOW [18:19] ah, fun, my patch makes AS segfault [18:19] seems like a good moment to go away for the weekend :P [18:19] have a good one everyone! [18:26] Laney, thanks, you too! [18:28] seb128, hey [18:29] seb128, how do i go about getting this blueprint turned into a uds session? https://blueprints.launchpad.net/webapps-applications/+spec/client-1311-webapps-sru-alternatives [19:01] chrisccoulson, where do you look to know if a firefox bug is known upstream? e.g https://launchpadlibrarian.net/145559707/Stacktrace.txt [19:02] chrisccoulson, is hunspell a distro stuff or upstream? [19:15] seb128, i guess have a look at https://crash-stats.mozilla.com/home/products/Firefox [19:15] we should really get this crash reporter enabled again [19:15] sil2100, do you still need that package sponsored? [19:16] sil2100, did you see my feedback on irc yesterday? [19:47] kenvandine: hi! Yes! I'll fix that override and look at that lintian error :) [19:47] kenvandine: are those the only problems with the package? [19:47] sil2100, yes [20:10] robru: http://calypso.cyphermox.net/~mtrudel/component_mismatches_ppa.png [20:10] ^ ubuntu-download-manager [20:11] kenvandine, ^^ [20:11] not a whole lot to do [20:11] cyphermox, what's the deal with webtest? doesn't look connected [20:11] or at least, much less than the prvious [20:11] no, it's not connected [20:11] it's an irrelevant other thing [20:12] cyphermox, also hunspell? it's not u-d-m pulling that in? [20:12] nope [20:12] just look at u-d-m and anything below it [20:12] so just qtsystems then? and it pulls nothing else in? can it really be that easy? [20:13] robru: compare to http://people.canonical.com/~ubuntu-archive/component-mismatches.svg [20:13] yes [20:13] cyphermox, oh, ok. great [20:13] you should still take a look to make sure qtsystems has all its build-depends in main [20:14] but the diagram should be fairly good [20:14] cyphermox, bah! i thought this tool did that? [20:14] it does [20:14] it doesn't hurt to double check though ;) [20:16] kenvandine: ok, I fixed the issues and pushed to the branch I mentioned earlier :) [20:16] kenvandine: the capnproto one [20:16] cyphermox, too good to be true: http://paste.ubuntu.com/6384111/ [20:17] I think you'll find all those to be binary packages that are built from qtsystems. [20:18] hmmmm [20:24] Laney, can you check the mozjs17 "js" shell on powerpc? https://bugzilla.gnome.org/show_bug.cgi?id=711519 [20:24] Gnome bug 711519 in general "jsunit tests fail on powerpc" [Normal,Unconfirmed]