[01:20] robert_ancell: yeah. And mvo said he'd try to fast track reviews of any followups I can get ready in time for 2.28 [01:33] jamesh, thinking about the best way to migrate snapd-glib. I guess there's no way to know if snapd supports this unless we get it to advertise it. Perhaps we should make a patch to add a polkit flat to /v2/system-info [01:34] actually we could just check the version field. [01:34] the current version is "2.27.3+17.10". I guess a heuristic that looks for >2.28 would work [01:35] though a flag would be less ambigious [01:37] robert_ancell: on Ubuntu at least, we could just do Requires: snapd >= 2.28 [01:37] in the packaging [01:37] sure, and just encourage other snapd-glib consumers to do the same [01:38] I'll mark the snapd-login-service methods as deprecated to prompt people to switch over. [01:56] jamesh: the On/Off switch for g-c-c's Connectivity Check doesn't display right when switched on/off [01:57] but hey, it's Feature Freeze so pushing anyway ;) [01:59] I remember that Feature Freeze is the day to break things :) [02:07] jbicha: it should be tied to the D-Bus property change notification. Let's see. [02:07] i.e. toggle switch issues Set() call, and the change signal updates the switch position [02:15] jamesh, did you get gnome git access? [02:16] robert_ancell: not yet [02:16] jamesh, where's your classic mode branches stored? [02:19] robert_ancell: this has the two patches I put in bugzilla: https://code.launchpad.net/~jamesh/gnome-software/+git/gnome-software/+ref/classic-snap-install-upstream [02:20] which branch should I rebase them onto? [02:21] jamesh, I'll push them into wip/ubuntu-master and then we can try and move them to master from there [02:21] okay [02:23] jamesh, once you get your git access the process is 1. in wip/james/ for stuff as it's being worked on, 2. in wip/ubuntu-master (artful), wip/ubuntu-3-22 (zesty) and wip-ubuntu-3-20 (xenial) when it's good enough for Ubuntu, 3. work on getting it into master when upstream is OK. [02:23] Ideally we go steps 1 - 3 - 2 but depends how much discussion is needed / priority in getting feature into Ubuntu [02:34] jamesh, someone OK'd the placement of that classic warning right? It's not even visible when I browse atom (you have to scroll down) [02:36] robert_ancell: no. It hasn't had any design review [02:36] the same problem would be true for the kudo icons if we were showing them [02:38] jamesh, "if we were showing them" :) [02:38] jamesh, I'll release it to artful, and hopefully we can get some feedback if it needs moving. [02:39] jamesh, Richard said he'd got some data at some point about what parts users were interacting with and it gets orders of magnitides less the futher you go down the page. [02:39] jamesh, I think we should move it to beside the install button. [02:40] * robert_ancell heads out, will be back later [02:40] robert_ancell: from a user's point of view, I'd think it'd make sense to show the warning for strict confined snaps that connect to home too, but it isn't clear how to do that with current snapd [02:40] true [02:41] we don't know until it's installed... [02:41] it also doesn't help that we're mixing a small number of snaps with a large number of debs in the same search results [02:41] where the debs are effectively unsandboxed [02:42] maybe it would make sense to also show the warning for debs that come from non-Ubuntu sources (PPAs and other archives) [03:07] jamesh: that sounds like a good idea [05:18] good morning [05:20] Hi jibel [05:20] Hey duflu [05:21] hey jibel, duflu [05:21] good morning desktoppers [05:22] Morning oSoMoN [05:33] morning oSoMoN [05:59] Wow. Some G-apps touch eight graphics APIs; GTK, GDK, clutter, cairo, cogl, pixman, GL, X11 [06:00] That's more than I thought [08:03] moin [08:08] Laney, hi [08:08] Laney, about bug 1712581 [08:08] bug 1712581 in ubiquity (Ubuntu) "Cannot execute oem-config-prepare - application is not trusted" [Undecided,New] https://launchpad.net/bugs/1712581 [08:09] Laney, if I set the attribute on the file in /target at the end of the installation it should be preserved after it's unmounted and rebooted right? [08:09] Morning all [08:09] Hey davmor2 long time no see :) [08:10] jibel: I've been here just not saying hello :D [08:10] hi jibel, ca va? [08:10] Laney, ça va bien merci et toi? [08:11] jibel: It saves to some database, so if you execute it in the right context it should be [08:11] that's why we have to sudo to the user, start a bus and all that random stuff [08:11] Laney, ah, maybe I did it with the wrong user. I'll have another look [08:12] je vais bien! [08:12] :) [08:12] * Laney tries to drag up the 10 french lessons from 15 years ago [08:12] * Laney did german in school :p [08:13] jibel: got a link to the code for this thing? [08:13] maybe I can advise quickly [08:15] Laney, http://bazaar.launchpad.net/~ubuntu-installer/ubiquity/trunk/view/head:/scripts/plugininstall.py#L1190 [08:15] Laney, with a comment from 2007 from Colin :) [08:16] ah yes [08:16] * Laney found some 2007 cjwatson code earlier this week too [08:16] maybe it's time to light the chemical fire :D [08:16] looks like you probably basically translate that same thing [08:17] guessing chrex is execute in chroot [08:17] put it at 1215 [08:17] Laney, yeah that's what I did [08:18] Laney, but it's likely executed with the wrong user [08:18] not the oem user [08:19] 'sudo', '-i', '-u', 'oem', 'dbus-run-session', '--', 'gio', 'set', '/home/oem/Desktop/%s' % desktop_base, 'metadata::trusted', 'true' [08:19] or something [08:19] yup [08:19] that should run it as oem... [08:19] you can do `gio list ' to see the properties [08:19] or is it info? [08:20] info [08:20] Laney, i'll figure it out [08:22] okey doke [08:37] pull request sent through to turn on polkit auth for install/remove of snaps [08:37] lets see what happens now [08:38] getting interesting :-) [08:41] Laney, it works, i'll submit a patch [08:41] ^_^ [08:41] jibel the developer [08:41] hi seb128 [08:42] what up [08:42] hey Laney! how are you today? [08:43] Laney, do you know what calls finish-install.d/07oem-config-user ? the change must go there too I suppose [08:43] I had a nice day off and some good sleep yesterday :-) [08:43] seb128: good! nice climbing yesterday, did some planting and harvesting this morning [08:43] we had some big potatoes growing at home [08:43] did a bit of work in the afternoon and before going to bed so swap some hours out today as well [08:43] who needs anallotment [08:43] oh did you! [08:43] you minx [08:44] hehe [08:44] jibel: not sure [08:44] you see an additional problem? [08:44] Laney, you are not going to London with didrocks&co after all? [08:44] Laney, no but it's exactly the same code in shell [08:45] jibel: oh ok, when's that used? [08:45] I guess that is the question you're asking :P [08:45] seb128: nah, too much travelling [08:45] Laney, I don't know, hence my question. [08:45] jibel: I would ask xnox [08:46] xnox, in ubiquity there is finish-install.d/07oem-config-user, do you know what calls it? [09:12] Laney: hey, I have a new patch, this time I followed what upstream recommended: https://code.launchpad.net/~albertomilone/gdm/lp1697882/+merge/326426 [09:13] tseliot: oh, nice, thanks for continuing to work on that [09:13] will look later on [09:14] Laney: thanks [09:27] tseliot: maybe you can ping the person you talked to to review in the meantime ;-) [09:28] Laney: it was you and Daniel van Vugt, but I don't know what his nickname is [09:29] ah, duflu ^^ [09:29] oho [09:29] we're not upstream :P [09:29] not yet, but I spoke with upstream until yesterday evening [09:30] ah yeah, that's who I meant to ping [09:30] Ray Strode. He had a look at my patch yesterday [09:31] cool [09:32] tseliot, thanks for that. Although I don't use Nvidia most of the time. Was only the frontend person triaging those bugs [09:32] duflu: ok, no problem [09:33] Laney: he said he wants us to discuss the upstreaming process in the bug report. But I did exactly what he recommended. He would also like me to add an IgnoreXorg option to GDM, but that's a separate upstream discussion. [09:34] I think avoiding [09:34] *preventing users from crashing things is the priority for us now [09:35] why can't nvidia/kms users use x11? [09:35] if you log in with nvidia + KMS and X11 all you will see is a dead terminal [09:35] we don't enable kms by default though? [09:35] seb128: a limitation in NVIDIA's memory allocator [09:35] nope [09:36] k [09:36] but if users enable KMS, they should be able to try Wayland without experiencing that problem [09:38] right [09:38] it sucks a bit that they can't log back into an x11 session if they need to [09:39] yes, they will have to disable, KMS, and reboot [09:39] but if it only impacts users who know what they are doing and opted in to enable kms so no big deal [09:39] exactly [09:39] I see it as a technology preview [09:39] as XWayland won't work either for now [09:40] but at least users get to experience pure Wayland with NVIDIA [09:54] hey Trevino, I'm working on bug 1700100 and looking about adding a "darker border" as you suggested, but not satisfied with the result (or I misunderstood something) dim-label has already some default decoration, which seems to work fine in other application (e.g. Nautilus), the problem with Boxes is that the parent background is too dark, so what about adding some light text shadows instead? Something like this http://imgur.com/a/eXpaW [09:54] bug 1700100 in ubuntu-themes (Ubuntu) "Dimmed text is very hard to see" [Undecided,In progress] https://launchpad.net/bugs/1700100 [09:55] also, hi all :) [10:03] oops, wrong nick. Trevinho ^ [10:03] clobrano: oh yeah... I like this alot [10:05] clobrano: the first one is ok to me [10:05] as it has something, but not too strong [10:05] go for it :) [10:05] Trevinho, ok, the first one is actually the current proposal :D, I don't have to do anything [10:06] clobrano: oh ahahah, :-D [10:06] :D [10:06] clobrano: anyway, that could be applied to other apps no? [10:06] firefox for example... [10:06] what others are? [10:07] Trevinho, firefox doesn't seem affected by this change. I suspect it's using a different class [10:07] Trevinho, and thunderbird as well. I should look into their code to see what are they using [10:10] #launchpaddown? [10:10] someone on X & current artful want to try a webkit application? e.g. devhelp or epiphany [10:10] * Laney gets an X error [10:11] Laney, devhelp works [10:11] whyyyyyyyyy [10:12] * Laney turns it off and on again™ [10:14] Laney: you've to pay a royalty do do that! [10:15] Laney, it works BUT ... /usr/lib/x86_64-linux-gnu/webkit2gtk-4.0/WebKitWebProcess crashed [10:15] ah wait [10:15] click on an entry to actually view the contents of a devhelp document [10:15] it doesn't crash, you just don't see it [10:15] well yeah, maybe the backgroudn process crashes but devhelp doesn't [10:15] * Laney give Trevinho 1BTC [10:16] that's actually quite a lot isn't it? [10:16] Laney: I'm not M$ [10:16] eh, yeah... I can get them then [10:16] Laney, as I said devhelp works fine [10:16] Laney, I can see the content of the documents [10:16] ok, I realised I forgot to tell you to click into something [10:16] but I've this webkit crash in /var/crash [10:16] anyways it works for me after a restart :( [10:17] wonder what the crash you had was [10:17] I'm uploading it and will try again [10:22] Laney, it crashes on exit [10:22] https://errors.ubuntu.com/oops/6b38c1d2-88b5-11e7-be89-fa163ed44aae [10:22] intriguing [10:22] * Laney needs a 2fa token to see that and none are nearby /o\ [10:23] k, presumably that still needs retracing [10:24] yes, I'll let you know when its done [10:25] * jibel -> lunch [10:29] hey seb128 [10:30] hey andyrock [10:30] how are you? [10:30] can I use git ubuntu clone for update-manager instead of debdiff? [10:30] just curious [10:30] I think it's still in bzr ? [10:30] talk to bdmurray on #ubuntu-devel [10:30] kk [10:49] o/ all [10:50] Could I get a review on this from a GNOME guy/gal? https://bugs.launchpad.net/ubuntu/+source/gnome-desktop3/+bug/1703690 [10:50] Ubuntu bug 1703690 in gnome-screensaver (Ubuntu) "Add support for Budgie Desktop using GNOME Screensaver" [Undecided,In progress] [10:51] Feature Freeze is today :) [10:54] tsimonq2: ok, I don't understand the gnome-desktop3 patch so please can you explain what that's trying to do more clearly? [10:55] two things on the gnome-screensaver one: in_desktop should probably use g_strv_contains instead of iterating the list and I think you should find (or create) a common file to contain that utility function rather than defining it multiple times [10:57] Laney: Ok, fossfreedom can answer those questions better, I just wanted to get some eyes on it. ;) [10:59] Laney, o/ I can help with any questions. [11:03] For gnome-screensaver it is carrying a patch which reinstates a gsettings key. Think its a Unity thing but not sure. gnome-screensaver calls a gnome-desktop3 function. the function assumes the gsettings path is valid and all keys it interrogates are there. For GNOME3 the "Unity" key doesnt exist ... and thus segfaults [11:04] to be honest if the gnome-screensaver patch that reinstates that Unity key can be got rid of, the gnome-desktop3 patch is not needed [11:04] which key is that? [11:06] btw, g_object_get () is more convenient when you're using C [11:06] you can get strings out directly [11:06] "draw-background" in path org.gnome.desktop.background [11:07] draw-background doesnt exist in gnome3 - org.gnome.desktop.screensaver [11:08] it's in org.gnome.desktop.background here [11:08] I feel like I'm confused about this [11:09] ok - when you change the screensaver background in gnome-shell it writes the background file chosen into path org.gnome.desktop.screensaver [11:10] good morning [11:10] gnome-screensaver was written in the days when the screensaver background was previously written to org.gnome.desktop.background [11:10] o/ jbicha [11:10] so gnome-screensaver never displays the chosen gnome3 screensaver background [11:11] Laney: should I go ahead and file FFe bugs for gnome-shell and gnome-control-center? [11:12] jbicha: Sure [11:12] I don't know what to do about the armhf thing [11:13] I understand, and didrocks is busy with the F&F Sprint in London for several days so probably next week to figure out what we want to do [11:15] right [11:15] if we want we can promote a PPA in the weekly newletter or something [11:15] fossfreedom: so someone is calling load_from_preferences with a different schema now? [11:15] ouch another theme glitch https://usercontent.irccloud-cdn.com/file/gbrMS0ER/ [11:16] even when not focused.. [11:16] clobrano: would you like to have a look on that too? :-) [11:16] I think I'd make a new GSettings in there to look at the old key [11:17] Laney yeah - that patch proposed checks if "budgie-desktop" and calls the method with the newer schema. load_from_preferences assumes all the keys it checks exists in the schema. [11:18] sorry - gnome-desktop3 itself has a patch that reinstates that "draw-background" key [11:19] fossfreedom: make the is_enabled check use a new GSettings, like GSettings *background_settings; background_settings = g_settings_new ("org.gnome.desktop.background"); bg->is_enabled = g_settings_get_boolean (background_settings, BG_KEY_DRAW_BACKGROUND); g_object_unref (background_settings); [11:21] Laney, the webkit crash is bug 1711838 [11:21] bug 1711838 in webkit2gtk (Ubuntu) "WebKitWebProcess crashed with SIGSEGV in WebCore::GLContextGLX::~GLContextGLX()" [Medium,Confirmed] https://launchpad.net/bugs/1711838 [11:22] thx [11:22] Laney, k - sure. That's will simply matters. good lateral thinking! [11:22] simplify [11:22] * Laney is a simple person [11:25] Trevinho: sure :) is the problem that white border? [11:25] c-lobrano: yeah, and when unfocused it's just squared instead of rounded [11:25] why are there two schemas with a lot of the same keys? [11:26] Trevinho: I see, I'll look at that [11:27] Laney, I can only assume upstream GNOME changed to a newer schema at some-point. Maybe the old schema will at some-point will be dropped. speculating. [11:27] c-lobrano: thanks a lot [11:28] Laney, or logically - one schema specifically for gnome shell screensaver background... one schema for the desktop background [11:29] yeh maybe [12:00] Trevinho: switches look bad too, same thing? [12:01] there's a more obvious one but I can't remember where it is, so - gnome-tweak-tool -> keyboard & mouse -> click the thing to change the compose key [12:01] c-lobrano: ^ :) [12:01] in fact switches just in the UI could look a bit better too I think [12:01] the orange sort of bleeds out [12:01] this is my fault of course when doing the theme rewrite :-) [12:04] * c-lobrano is updating the todo list :) [12:05] thanks c-lobrano! [12:05] the theme is a hot potato [12:05] * Laney held it for a bit and then passed it on to Trevinho [12:06] Ahaha [12:07] brains are weird [12:07] "oh yeah I'll start a VM to test this thing" [12:07] hit super, type "devhelp" instead of "virt"... [12:07] what is that about [13:25] jbicha, those MIRs can be changed to FFes but we don't have much control on them, unsure which ones would be worth pushing for this cycle, some feel not important [13:29] jbicha, do you have branches/ongoing work for g-c-c and updating the patches or do you need help? [13:30] jbicha: IIRC, you told that the power button was suppose to be fixed to trigger a dialog [13:30] that was some weeks ago, is it fixed? [13:30] didrocks, hey! how is London? [13:31] seb128: same temperature outside that at 7am in Lyon :) [13:31] otherwise, not rainy, so good :) [13:31] haha [13:32] didrocks, https://git.gnome.org/browse/gnome-control-center/commit/?id=62c86c7 [13:32] didrocks, it's part of 3.25 which we didn't get yet/is going to be a ffe [13:32] ah ok! what is the dialog supposed to look like, do you know how I can try it? [13:33] no, the bug has no screenshot [13:33] mpt in software-properties enabling/disabling canonical livepatch can requires some time (as it can involve downloading/installing the snap, networking, etc.) [13:33] I guess it's easy enough to install the new g-s-d/g-c-c from the ppa though [13:33] mtp, atm I'm showing a spinner next to the checkbox [13:34] mtp, but this is not specified on the design at: https://wiki.ubuntu.com/SoftwareUpdates?action=recall&rev=221 [13:34] seb128: it's g-s-d providing this dialog I guess? [13:34] mtp, is a spinner ok for you? or do you have any other solution? [13:35] didrocks, looks like if I read things correctly [13:37] oSoMoN, ricotz, hey, do you have any idea about http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html#libreofficehttp://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html#libreoffice ? [13:37] http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html#libreoffice [13:37] thanks seb128 [13:37] sorry copy/paste issue [13:37] didrocks, np! [13:38] didrocks, the new g-s-d is in artful-proposed but seems blocked there because gnome-session is missing on s390x [13:41] seb128, isn't that caused by poppler being blocked by qtbase-opensource-src ? [13:41] didrocks: the Power Off dialog is the same as when you press the Power button in the system status menu in GNOME Shell's top bar on the right [13:41] but the behavior for the hardware Power button is still Suspend by default [13:42] seb128, failing i386 tests are still caused the kernel/openjdk bug [13:42] but gnome-settings-daemon is built there [13:42] gnome-session* [13:42] it's part of gsd and gcc 3.26; it didn't seem important to backport since we're working on targeting g-c-c 3.26 for artful soon anyway [13:42] https://launchpad.net/ubuntu/artful/s390x/gnome-session-bin removed [13:43] was there a real reason for that? [13:43] seb128: I have done no work on porting those 2 disabled gcc patches, I need help with them, the current g-c-c work is in the GNOME3 Staging PPA [13:43] jbicha, do you have a vcs with your work in progress? what sort of issue do you have? [13:45] jbicha, ^ do you about gnome-session? [13:45] no vcs really right now, kinda waiting for git conversion to push more WIP branches [13:45] I don't see it mentioned in the changelog [13:46] Laney: oh, removing gnome-session-bin/s390x was likely a mistake as part of the gjs/s390x purge [13:46] you can push to lp:~user/g-c-c/work or something [13:46] is it possible to retrieve that binary or do we need to do a new upload? [13:46] copy it back [13:47] copy it from where? [13:47] sec [13:47] k [13:48] copy-package is for sources only right? [13:49] * seb128 let L_aney come up with the magic [13:49] I thought you could use it but I don't see the option right now [13:49] andyrock, mtp is the Media Transfer Protocol. :-) (I got that a lot during the Ubuntu Touch era) [13:49] give me a minute, need to share a screenshot to aday [13:49] mpt: ops sorry :D [13:50] jbicha, btw versions page updated in case you didn't notice [13:50] thanks [13:50] yw! [13:51] andyrock, is there any reason that you would need/want to wait for Livepatch to become active/inactive after toggling the checkbox? [13:52] (where by “you” I mean a user, not you the person implementing it!) [13:52] mpt: I mean I'm a user a click on "Use Canonical livepatch to increase..." [13:53] k, yeah, don't see the option [13:53] I get no feedback for the next 10 minutes [13:53] I'll try to click multiple times [13:54] jbicha, Laney, I think it needs a new upload then [13:54] andyrock, you don’t have to wait ten minutes before signing in, right? You can sign in immediately [13:54] just asking in #lp [13:54] not sign in... [13:54] see what the superiors say ;-) [13:54] mpt: but enabling livepatch can require some time [13:55] mpt: e.g. if it's not installed the snap it has to download it etc. [13:56] just make it async ... set the checkbox immediately ... if there is an error or other issue, uncheck it again and pop up a useful error message [13:56] andyrock, sure, but there’s no reason to sit there watching it. It’s like turning on backups: maybe it’s satisfying to watch the first backup in progress, but you don’t need to. [13:56] Yes, what ogra_ said ^^ [13:57] (much more clearly than I did) [13:57] (but i learned it all from years of working with you ;) ) [13:57] seb128, ricotz: there are two failures in i386 autopkgtests, one of them is the infamous kernel/openjdk bug, and the other one is new (and observed on s390x too): /usr/bin/ld: cannot find -ljawt [13:57] looking into that [13:59] https://launchpad.net/ubuntu/artful/s390x/gnome-session-bin it's coming back [13:59] btw was removing gdm3:s390x intentional? [13:59] * xnox had to tweak systemd autopkgtest [14:00] oSoMoN, yeah, looks like an incomplete linker line and -L/usr/lib/jvm/default-java/jre/lib/* is missing [14:05] kk [14:08] oSoMoN, e.g. https://paste.debian.net/plain/982856 [14:09] .. which obviously isn't working just like that [14:10] meh [14:12] not sure this restoring of gnome-session is going to work, it depends on gnome-shell which was removed as it depends on gjs which is broken [14:12] * Laney remembers this stuff from phone/upstart days :( [14:13] ricotz, do you mean you've tested that patch and it's not working? [14:14] oSoMoN, I pointed out the place where it needs to be fixed ;) [14:15] basically adding the proper -L... here will fix it [14:15] strange that amd64 is happy though [14:15] ricotz, yes, that's strange [14:15] ricotz, replacing -ljawt with $(JAWTLIB) as you suggested should work [14:19] Laney: I think gnome-session-bin does not depend on gnome-shell [14:20] seb128: preparing the patch suggested by ricotz [14:21] seb128: https://code.launchpad.net/~ubuntu-desktop/gnome-control-center/326 [14:21] gnome-session does [14:22] oSoMoN, I am not so sure though [14:23] oSoMoN, the configure call has nothing to do with that [14:24] ricotz, the configure call is what sets JAWTLIB [14:24] and I'm seeing that in the build log: [14:24] checking for JAWT lib... -L/usr/lib/jvm/default-java/jre/lib/i386 -ljawt [14:24] but it is not part of the odk test [14:27] oSoMoN, better try https://paste.debian.net/plain/982862 [14:27] ah no [14:28] oSoMoN, JAVA_PROC_TYPE is wrong [14:31] seb128: do you want gnome-desktop3 3.25.91 in artful now with bubblewrap disabled (which breaks the thumbnailer anyway) or should we wait and discuss more later? [14:31] huh indeed, it's set to amd64 [14:33] Trevinho: from where this come from? https://usercontent.irccloud-cdn.com/file/gbrMS0ER/ [14:33] gnome-session/s390x unsatisfiable Depends: gnome-shell (>= 3.19) [14:33] ubuntu-session/s390x unsatisfiable Depends: gnome-shell (>= 3.24.3-0ubuntu2) [14:33] clobrano: evince [14:33] (y) [14:37] Trevinho: weird, I don't have those plus and minus buttons, but only something with a similar problem in page number [14:37] clobrano: oh sorry not evince... it was image viewer [14:37] if you open a multiple page document in evince [14:38] the x of y thing at the left of the header bar is bad too [14:41] Laney: I'm confused too because gnome-session and ubuntu-session aren't published for s390x in artful [14:41] https://launchpad.net/ubuntu/artful/s390x/gnome-session [14:41] sure they are [14:43] maybe my rmadison is a bit behind then [14:44] oh, artful, not artful-proposed [14:44] well, britney doesn't care about that [14:44] Trevinho, Laney: are you on artful? On zesty and this light-themes package 16.10+17.04.20170406-0ubuntu1 I can't reproduce the issue on image viewer: http://imgur.com/a/xmjtT [14:44] you don't get to build uninstallable packages [14:45] well, even with latest light-themes from bzr [14:46] maybe remove gnome-session and ubuntu-session binaries only and arch-restrict those in the source so they don't come back [14:46] clobrano: yeh, on artful/gnome-shell [14:46] * clobrano needs a VM then :) [14:49] Laney: do you want me to try re-uploading without building those for s390x then? [14:50] maybe we can get seb128 or someone to try removing them [14:50] don't actually have to upload as long as the next one doesn't build them [14:50] or fails to build them [14:50] that's an interesting distinction [14:50] * Laney fluffles xnox [14:51] ooooh so cute https://vignette4.wikia.nocookie.net/wallaceandgromit/images/2/29/04007e876c0b46ccaf970c471c7cf4f9a836b0c5_b.jpg/revision/latest/scale-to-width-down/299?cb=20170605133538 [14:51] thanks [14:51] these stupid per arch removals are always annoying [14:51] in 99999 years when all is ash we'll wish we just fixed gjs/s390x [14:52] Add a Build-Depends on gjs then? [14:52] that was the phone solution [14:52] it became very annoying [14:53] ricotz, if JAVA_PROC_TYPE is wrong, it means that PROCTYPE is wrong, i.e. that PLATFORMID is wrong, i.e. that host_cpu is wrong, i.e. that AC_CANONICAL_HOST produces an unexpected result [14:53] ricotz: I didn't get around to working with vala this week, you didn't happen to do an archive rebuild of the new version right? [14:53] this is going down way too many layers [14:54] jbicha, no, just pushed the packages to the vala daily ppa as I mentioned [14:55] oSoMoN, right, but I would assume the build-hosts are reporting things correctly and therefore this build sources something hardcoded [14:55] ricotz: are you ok with waiting until 18.04 for new vala then? [14:55] oSoMoN, note this is part not driven by configure [14:56] oSoMoN, I am leaning to that config_host.mk(.in) is to blame? [14:59] jbicha, you know my anwer ;) [14:59]