[05:17] hello [05:37] Good morning === ogra_ is now known as ogra [08:03] oh god! [08:05] Laney is leaving IRC during the night now? [08:06] Subject: Your BitFolk VPS ("cripps") needs to be rebooted for critical maintenance [08:06] I got that a few days ago [08:06] but forgot :( [08:08] haha [08:09] ho hum [08:09] how's it going? [08:09] good! [08:10] it's friday right? ;-) [08:10] I was a bit down yesterday from daily iso testing and xenial bugs [08:10] it would seem so [08:11] but a good night sleep and I'm better today ;-) [08:11] how is it going for you? [08:20] pretty good! [08:20] we started watching stranger things last night [08:20] morning all, having internet problems today [08:20] joining everyone else in the world [08:20] creepy [08:20] hi willcooke [08:20] are you on DSL? [08:20] hey willcooke [08:21] yeah, dsl [08:21] hum [08:21] shucks [08:22] Trevinho, pitti, Laney, unity is blocked in yakkety-proposed due to unity8 autopkgt but those seem to hit a enospace, so infra issue? [08:23] bah, same of n-m-a/cinnamon-control-center it seems [08:23] it was https://bugs.launchpad.net/ubuntu/+source/util-linux/+bug/1619285 [08:23] Launchpad bug 1619285 in util-linux (Ubuntu) "cc_growpart fails on yakkety" [Critical,Confirmed] [08:23] i'm not sure if you need new images or if just retrying is enough [08:23] "was"? [08:24] so I just need a retry? [08:24] hum, k [08:24] it's fix released [08:24] so that should mean was [08:24] good morning Laney and seb128! [08:24] ahoy pitti! [08:25] seb128: ENOSPC is known, bug 1587188 [08:25] bug 1587971 in cloud-utils (Ubuntu) "duplicate for #1587188 recent versions of util-linux do not like how growpart calls partx" [Medium,Fix released] https://launchpad.net/bugs/1587971 [08:25] the fix/workaround just landed, now we need new cloud images [08:25] hey pitti, how are you? [08:25] Laney: btw, I eliminated browse.cgi, just rolling out [08:25] Laney: the trick was to put "ScriptAlias /" last in the list [08:25] seb128: I'm great, thanks! [08:27] pitti: excellent [08:41] seb128: nooo... Again?! :-( [08:41] Trevinho, hey [08:42] Landing these days is getting pretty annoying [08:42] Hey seb128 [08:42] Trevinho, well, that's another bug which is fix commited, see backlog [08:42] infra/yakkety issue [08:43] seb128: ah, I seee... In landing queue too? [08:44] Trevinho, it's landed in yakkety if I understood correctly, they just need new images to be built [08:44] unsure who/when for those though [08:44] Mh, I see [08:45] So... We'll wait for landing unity7 too... [08:46] I decided to use a different silo for systemd since it would have been too complicated to do everything.... Anyway, code wise things are ready now. I only need to get trunk updated [08:46] k [08:52] Trevinho: \o/ great news! thanks [08:59] seb128: thank you, :) [08:59] happyaron, hey, yw! [09:00] happyaron, it's currently stucked in proposed due to the infra issue discussed a bit earlier but that should clear off at some point (today?) [09:02] I should think pitti is rebuilding the image and then will retry everything [09:03] Laney: I can't rebuild cloud images [09:03] let's see, that part of the infra is non obvious to me [09:04] oh well, maybe rebuilding an adt one will do it, it might rerun cloud-init [09:04] let's try, it can hardly hurt [09:04] is there anyone we can ping otherwise? [09:04] or is that just sit there and wait for somebody to do $things [09:05] utlemming isn't online [09:05] otherwise the cron job will kick in by tomorrow [09:05] k [09:06] pitti: utlemming no longer works for us; you need one of (me|rcj|philroche|josvaz). :) [09:06] Laney, I think you worked a bit on that before, is there an easy way to start manually the slideshow part of ubiquity? [09:07] hey Odd_Bloke :-) [09:07] Odd_Bloke, seems we need you! :p [09:07] What's up? [09:07] seb128: I commented on the bug a few minutes ago with a diff [09:08] the answer is yes [09:08] Odd_Bloke, there was a cloud-utils bug which impact autopkgtest infra and just got fixed, seems we need new cloud images to be built now [09:08] Odd_Bloke: oh, hey! [09:09] how very Odd of you to turn up just at the right time [09:09] Odd_Bloke: after that cloud-utils fix is in, we need new cloud images that contain it, so that the resizing failure gets fixed [09:09] Odd_Bloke: do you have the powers to trigger new ones? [09:09] Laney, ah, great, one less things on my list then, thanks ;-) [09:09] seb128: still need to be fixed ... [09:09] ah [09:09] I stole a fix from fedora in ppa:laney/ppa but not that confident it will work [09:09] I hadn't open the bug yet [09:10] pitti: I do have the powers. [09:10] I though your diff was a fix :p [09:10] no [09:10] it just shows you how to start the slideshow [09:10] thanks [09:10] Laney: Because everyone still thinks utlemming is responsible for this, I have mentions of him highlighted. :p [09:10] would prefer it if the updater of webkit2gtk would look at it :/ [09:10] Odd_Bloke: ah, very conscientious [09:10] Laney, yeah, me too [09:11] * Laney goes to work on gstreamer srus while that builds [09:11] well, coffee first [09:11] jbicha, hey, saw bug #1618956? seems a side effect of the webkit2gtk update [09:11] bug 1618956 in webkit2gtk (Ubuntu) "Slideshow blank during live install" [High,Triaged] https://launchpad.net/bugs/1618956 [09:12] pitti: seb128: I saw seb128 say "now" so I've triggered builds, but then I saw pitti say "after that cloud-utils fix is in"; should I cancel this build, or is 0.29-0ubuntu3 the version we want? [09:13] Odd_Bloke: 0ubuntu3 is the version, so please keep it running [09:13] it landed a few hours ago already [09:14] Cool, that new image will filter out in due course. :) [09:14] Odd_Bloke: but it breaks autopkgtests left and right, hence the urgency; thanks, appreciated! [09:14] isn't cloud-init run every bood anyway? [09:14] boot [09:14] Laney: I'm not sure that it re-runs the growpart bit on every boot [09:15] Laney: so it's too late on instantiation (growpart fails, then you dist-upgrade); but building new adt images from the cloud ones could work, not sure [09:15] hm [09:15] but you don't know what size of machine it's going to be run on? [09:15] but if we get new cloud images anyway, I'll just build autopkgtest images from those and then it should work without much experimentation [09:15] Laney: hm, good point [09:15] ok, just trying intuition but not sure it's right :) [09:16] so rebuilding autopkgtest images should work indeed [09:16] assuming you dist-upgrade in your script [09:16] yes [09:16] oh, and scalingstack lags behind in cloud imports too, so let's just try that [09:17] can't hurt that much to try anyway [09:23] oh dear, the infra got Qted again [09:23] :-/ [09:27] /dev/vda1 20G 1.2G 19G 6% / [09:28] \o/ [09:28] Laney, seb128: ça march bien à nouveau ! [09:28] still need to wait for lgw to finish [09:29] * Laney cat /dev/urandom > /dev/vda1 in celebration [09:29] pitti, super! [09:29] still waiting on lgw to finish building, then I'll do a mass retry [09:29] (for ubuntu) [09:30] thanks! [09:44] hi all! [09:44] someone having pkcon errors in unity8 on lapto? [09:44] hey vigo, try #ubuntu-unity about unity8 [09:44] I always get the same error with great thanks seb128 ;) [09:45] hey hikiko did you see the remaining fix to do in https://code.launchpad.net/~hikiko/compiz/compiz.fix-expo-gap-bug/+merge/304076 ? [09:47] Trevinho, sorry no, I was doing something else [09:47] hikiko: I've seen the compiz move outline... [09:47] so you mean I have to add a gap on top? [09:47] hikiko: yeah, there's something weird there, see the vid... The window has to consider the panel (so workarea.top()) [09:48] yeah expo doesn't use workareas though [09:48] it needs a yoffset I guess [09:48] hikiko: I was wondering if havin a texture to move, instead than just an outline would be acceptable too... Like just using the texture of the window, without being animated or... LIke a a static screenshot. [09:49] well the window is a quad with a texture [09:49] hikiko: I mean, it does... Well, wether it's a top gap or using the workarea is the same... In normal case (left launcher) it takes account of it. [09:49] oh [09:50] Trevinho, the expo plugin didn't take into account any offsets from the beginning [09:50] I mean before we move the launcher to the bottom [09:50] no [09:50] it had some top gap, isn't it? [09:50] it had a left gap [09:50] and on the top [09:50] it had a buggy animation [09:50] wrong scaling [09:50] optionGetYOffset () too [09:51] which is something should be take in account together with bottom gap [09:51] I'll fix it, it's not just the offset [09:51] I have to change the translation and the scaling [09:52] I just say that before it wasn't correct because it was taking some offset or sth into account it just had a scaling effect that was hiding the problem, if I recall correctly during the animation the top gap was changing size [09:53] Ah, I see [09:53] and the final position [09:53] was the one we have now [09:53] but it doesn't look nice [09:53] so I am going to fix it [09:53] you are right on that [10:28] Trevinho, do we set a y_offset for expo in unityshell? [10:31] I think we don't [10:31] the yoffset is the default (24) [10:31] if I take it into account it looks more awkward :p [10:35] what's the best way to restore unity settings in xenial? my wife's account has no sidebar launchers [10:35] but nautilus === hikiko is now known as hikiko|ln [11:04] anyone? unity-tweak-tool --reset didn't help [11:14] tjaalton: is that xenial? It should restore by itself if plugins are missing [11:14] (when updated) [11:15] tjaalton: /usr/lib/x86_64-linux-gnu/unity/unity-active-plugins-safety-check should do the trick [11:16] tjaalton: if it doesn't start, unity upstart log could help [11:16] going to lunch, in case I'll be back later [11:16] Trevinho: ok, thanks [11:22] safety check runs === hikiko|ln is now known as hikiko [11:36] Laney, seb128: I pushed a version of libreoffice bumped to 5.2.1~rc2, disabling the cups for the autopkgtests, the gtk2->gtk3 default move, google auth fix, and moving packagekit integration up to libreoffice-gnome so that it stays around for gtk3 to the libreoffice ppa: https://launchpad.net/~libreoffice/+archive/ubuntu/ppa/+sourcepub/6850844/+listing-archive-extra [11:37] Laney, seb128: but I will be at the LibreOffice conference next week, so I dont know how we will handle bumping this into yakkety. [11:37] Sweet5hark, nice, do you need sponsoring or want to get the ppa feedback first? [11:37] just upload now? [11:40] seb128: I can prepare a version now, but wont be around much next week if there are issues -- and 5.2.1 isnt named 5.2.1 final yet (but will be at the conf). [11:40] Sweet5hark, it's ok, yakkety is not stable yet so we can deal with issues the week after that [11:40] seb128: but yeah, will prep a uploadable package (removing the ~yakkety6 from version essentially). [11:41] or we can block it in proposed for a bit if you prefer [11:42] seb128: well, I did a update test from a vanilla daily image to the version in the ppa -- it worked and passed manual smoketesting. so lets go without blocking ... [11:42] Sweet5hark, +1 [11:56] Sweet5hark, seb128, +1 [13:09] qengho, morning! How goes the uploads for the api? Still on for today? [13:09] willcooke: For stable Ubuntu, yes. Y is still broken. I think it's gcc-6 now. [13:10] That sounds crazy, I know. [13:10] both firefox and chromium? [13:10] qengho, oki. seb128 can we still upload to X without it in Y? [13:11] we can upload, it might take some convincing talking to the SRU team on why they should accept a SRU before it hits unstable but I think it's an easy enough situation that it should be fine [13:11] though it's friday [13:12] unsure why we drifted so much to upload to xenial... [13:12] seb128: Cr is exempt from SRU. I would have only asked you to upload to Y. [13:12] you mean? [13:13] we need a new key in xenial to restore service no? [13:13] Firefox will probably (?!) need SRU. [13:13] It's in main. [13:13] and chromium? [13:14] sorry I don't understand [13:14] chromium and firefox in xenial use the abused key right? [13:14] so they need to be switched to use a new one [13:14] Yes. And yes. [13:14] which means they need a xenial upload [13:14] Yes. [13:14] k [13:14] I though you were saying chromium doesn't [13:15] so what's blocking those uploads? [13:15] toolchain didn't change, it's a "simple" key swap? [13:15] like change the key id, upload? [13:15] I'm sure I'm overlooking something and I'm about to learn what ;-) [13:16] Nothing is blocking trusty, xenial except that yakkety is crashy and I don't know why and I didn't want to upload out of synch. [13:17] what's the issue with the out of sync? [13:17] if anything it might be good because it might give us stats on different series [13:17] I guess nothing. :\ [13:17] k, so can we go ahead with the uploads of the series that are ready then please? [13:18] Okay, I'll ask chris to upload. Should be fast. [13:18] thanks [13:18] thx [13:29] qengho, hi, is this about the google-oauth-api key? [13:31] ricotz: er, maybe. We're changing Ubuntu's keys for google services because of abuse. [13:31] ricotz, are you the one abusing our keys? ;-) [13:32] I guess I will have to update the last firefox beta too then [13:33] qengho, https://bazaar.launchpad.net/~mozillateam/firefox/firefox-trunk.head/revision/2056 [13:33] ricotz: Oh, yes. https://code.launchpad.net/~cmiller/firefox/googapi [13:35] qengho, better target and base on https://code.launchpad.net/~mozillateam/firefox/firefox.yakkety [13:36] ricotz: I suspect only the api-key matters, fwiw. [13:36] qengho, while chris already pushed something regarding this better coordinate [13:40] qengho, yeah, so I guess updating debian/ga would suffice [13:55] Laney: hey, new systemd MP is building at https://requests.ci-train.ubuntu.com/#/ticket/1839, if you could give it a look... It's mostly tedg's one with some fixes on top. [13:55] Trevinho: okay, just going to lunch but will look after [13:55] Laney: sure, take your time... packages should be ready by then [13:57] Laney, enjoy! [13:57] Trevinho, unity migrated in yakkety, see no days or fiddling around to get it through this time ;-) [13:57] seb128: yeah, it was a nice surprise :-) [13:59] * Trevinho hunting that bitchy mosquito... [14:28] Laney: thank you for fixing the autopkgtest for gjs [15:21] jbicha: sure [15:21] my guess didn't fix webkit btw [15:22] :-/ [15:23] would be interesting to see if rebuilding the previous version gives something working or not [15:24] go for it [15:25] not today, that's going to take longer to build than an hour I think [15:25] but maybe jbicha wants to try it ;-) === mpt_ is now known as mpt [16:08] . [16:08] ? [16:09] fcsking vpn. takes 3 minutes after I post something for IRC to realize it needs to reconnect (and the messages in the mean time go to nirvana). Reposting ... [16:09] seb128: http://people.canonical.com/~bjoern/yakkety/5.2.1/libreoffice-l10n_5.2.1~rc2-0ubuntu1_source.changes http://people.canonical.com/~bjoern/yakkety/5.2.1/libreoffice_5.2.1~rc2-0ubuntu1_source.changes [16:10] seb128: note this switches to gtk3 by default and has the patch that makes the toolbars light and readable -- _iff_ the ubuntu-themes patch is merged too. Without that it will still run, but by default be rather ugly. L_aney has a merge request for that. the patched ubuntu-themes is also in the libreoffice ppa. [16:10] k [16:10] Sweet5hark, I'm going to sponsor that [16:10] then start calling it a week [16:10] seb128: thanks! *hugs* [16:11] * seb128 hugs Sweet5hark back [16:11] * Sweet5hark has the weekend an one day to prepare two talks and a keynote ... [16:12] where is the conf? how much time do you have in trains/planes to work on that? ;-) [16:25] Sweet5hark, k, sponsored [16:26] on that note calling it a week [16:26] have a nice w.e desktopers! [16:26] bye! [16:26] cya seb128 [16:26] happy weekend! [16:26] thanks, you too! [16:26] thx! bye all! [16:26] see ya Sweet5hark [16:53] urgh, having to turn the lights on already [16:54] wait until the heating starts coming on [16:55] still 6° to go for that here [16:55] * Laney should get some graphs on the go [16:55] :) [16:56] ximion: any more comments on the langpacks? [16:56] (hi!) [16:57] I found out about the meson submodule stuff yesterday [16:57] we should use that for libmo [16:57] if it works properly [16:57] implies I have to write a meson.build, but that should be fun [17:05] k, night! [17:08] night Laney [17:08] I'm off too, ta ta [17:08] bye Laney... [17:08] I was about to bother you, but... WE time :-) [17:21] Laney: I'm at Akademy right now, so not much time to review [17:21] and I think the code is ugly, but when the ugly bits get replaced with libmo soon, that should be fine [17:21] so, no big concerns for merging :) === sabdfl__ is now known as sabdfl