[00:02] *sits on Dalton* [00:05] But how? [00:05] Manually? What manages it? [00:31] the core infra is all containerized in a single server to isolate the envs from each other for cleanliness. All LXD containers are mirrored weekly to an LXD server I have here at home for backups. most of the deployed infra just runs itself [00:31] initial setup was done by yours truly and Simon for some of it [00:32] and by core infra we mean Phab, Weblate (translations), Discourse, CI (Jenkins), etc. [00:32] some stuff is still in Canonical's stacks like ISOs. [00:35] it's defined which kernel will be on 19.10? [00:36] ask the Kernel Team that one [00:36] not sure we reached kernel freeze yet - probably did but I dont have the freezes schedule right in front of me [00:41] mmm ubuntu wiki tells me that next kernel meeting will be tuesday 1oth nov. 2015.... [00:41] sorry, [00:41] 10th. [00:41] https://wiki.ubuntu.com/KernelTeam/Meeting [00:42] they lurk on IRC in #ubuntu-kernel but I need the Eoan dev schedule lol [00:42] phone is derping with the wiki [00:52] Feature Freeze: August 22, 2019 … Kernel Freeze: September 26, 2019 … Final Freeze: October 10, 2019 [00:53] from: https://www.omgubuntu.co.uk/2019/05/ubuntu-19-10-release-features [01:04] also from https://wiki.ubuntu.com/EoanErmine/ReleaseSchedule [01:04] which i trust more 😜 [01:15] wxl: I'm doing the meta update thing, so far so good. [13:05] and not, it takes tooooooo long,I went to bed and my system didn't woke up. Will try in another system tonight. [18:05] @HMollerCl i could run it in my container if you want. it's always running [18:06] would be nice [18:21] Eickmeyer: did Studio have a failure with updating apt-xapian-index on today's ISO? [18:21] wxl: I haven't seen a fail mail... [18:22] * Eickmeyer goes to check some schtuff [18:22] might be a timing thing. yours might build later [18:24] wxl: Probably. I'll let you know if I get a fail mail. As far as I can see, ours is still 20190625, which means it might still be building. Usually it's done by about Noon PDT. [18:24] kk [18:26] tl;dr Lubuntu ISO builds are failing because apt-xapian-index is depending on a deprecated python-apt feature that python-apt removed in the latest release. [18:26] i saw an email giving a heads up about the apt change a few weeks ago I think [18:27] how did all the others pass through today? [18:27] about that section being removed and might cause issues [18:27] wxl: it's Simon's fault? [18:27] We're the only one that depends on apt-xapian-index. [18:27] (Assuming it's only failing for us.) [18:28] Studio, Kylin, and Budgie are on yesterday's [18:28] Studio may not have a problem though [18:28] i *thought* others used it [18:28] i seem to remember kubuntu did but i could be wrong [18:29] post-malone? [18:31] wat? XD [18:31] oh [18:31] no [18:31] ProposedMigration [18:31] wxl: Also, ECHAN [18:32] no, i wanted to carry that over here :) [18:33] XD [18:38] do we need it, especially if it's problematic and orphaned? [18:39] No clue. [18:59] as it's a Recommends of libqapt I'd have thought it is only needed if there are QT based software search tools using libqapt, like qapt-deb-installer [19:05] We have qapt-deb-installer. I think we added it for Disco. [19:05] @kc2bez [ We have qapt-deb-installer. I think we added it for Disco.], yes, we did [19:06] cause otherwise ark wanted to open them [19:11] @HMollerCl doesn't discover do those now, too? [19:12] Discover should be able to as well. [19:12] But Discover... [19:12] discover is getting better soon [19:12] really there's very little wrong with it [19:12] the major problem is the lack of progress indicators and that's on the way [19:12] It is way better already. [19:13] this is the "slow" thing [19:13] it's a perception, not a reality [19:13] we don't think of synaptic, apt, etc. as slow because we see an indicator, but they take the same amount of time [19:13] let's just get rid of it and be done with it. anyone got a problem with that? [19:13] It does use some resources on launch but it settles down shortly after. [19:14] or maybe we wait to see how things settle out with upstream [19:14] thoughts, @tsimonq2 ? [19:15] wxl discover doesn't install cli apps from a gui [19:18] let me try discover again in my old machines. [19:18] try it with a cli app of some kind. tmux, let's say, @HMollerCl [19:18] oh i guess i need to upload lubuntu-meta [19:19] my guess is that discover will consume uneeded resources only for open a .deb, like refreshing cache and others. [19:19] right, @tsimonq2 ? [19:19] wxl: I don't quite have a preference. [19:19] if muon does, muon is a better option. [19:21] wxl: Yep, got a fail mail. You're not alone. [19:21] let's all blame Simon [19:21] because reasons. [19:21] * Eickmeyer throws @tsimonq2 under a bus [19:22] @teward001 [let's all blame Simon], we could also sue him XD [19:24] hey XD [19:24] I think you don't need to worry if the issue a FTBFS - is that the case? [19:24] no it's not TJ- [19:25] wxl: I thought it was due to apt-xapian-index failing due to python-apt changes? [19:26] it fails but not fails to build [19:26] wxl: sorry, fails *tests* [19:29] wxl: without seeing the fail logs hard to tell, but I suspect this commit addresses the issue: https://salsa.debian.org/apt-team/python-apt/commit/68227b7f16a8e3a2569368cabe2dff03468336b3 [19:30] TJ-: https://launchpadlibrarian.net/430708403/buildlog_ubuntu_eoan_amd64_lubuntu_BUILDING.txt.gz [19:31] don't think that's it [19:33] wxl: you're correct; thanks for the log. Here's the commit that broke it: https://salsa.debian.org/apt-team/python-apt/commit/fedd51be48be53d00d386c45eab8f63f462db202 [19:34] "* The `section` attribute has been removed from :class:`apt_pkg.Package`" [19:34] yep nice catch [19:34] Looks like you need to report a regression bug against that as it broke apt-xapian-index [19:35] I was looking through the python-apt mailing list but couldn't find any discussion on removing those features [19:35] there's this https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=931132 [19:35] Debian bug 931132 in apt-xapian-index "apt-xapian-index: please enable autopkgtests" [Important, Open] [19:35] and this https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=931133 [19:35] Debian bug 931133 in apt-xapian-index "apt-xapian-index: unusable after recent python-apt update due to usage of deprecated Package.section attribute" [Important, Open] [19:36] which should, in reverse order, fix the problem and keep it from happening again [19:36] wxl: apt-xapian-index has been removed from Debian experimental [19:38] Removal from Debian Experimental means nothing. [19:39] Debian Experimental is just a test pocket. [19:39] tsimonq2: sorry, I got myself confused here [19:39] ah :) [19:39] np [19:39] * TJ- has a fever today and is quite spaced out [19:40] I think I need to turn off this flourescent green colour theme [19:42] I suppose the question is, how long will it take for the fix to percolate through [19:43] looks like there's a bugette in Setting up uuid-runtime (2.33.1-0.1ubuntu2) .. too [19:44] it's creating the group before the user and causing "Warning: The home dir /run/uuidd you specified can't be accessed: No such file or directory" [19:44] or maybe not; that's weird! "Not creating home directory `/run/uuidd'." [19:48] @tsimonq2: um. i'm confused. it doesn't seem that when i build lubuntu-meta it GPG signs anything, though afaik you need to force debuild to NOT do so. i just did the same thing on libfm-qt6 and all is well. but of course dput gets unhappy without a signature. catch-22? [19:50] alias "cd.."="cd .." [19:50] don't tell me you wish you had it at least at one point XD [19:52] wxl: debuild -S -d *-us -uc* [19:52] Those two flags go without signing it. [19:52] yeah i *want* to sign it [19:53] However, if you're sponsoring an upload, you still need to run debsign -kMYKEYIDHERE /path/to/source.changes [19:53] so `debuild -S` should just do it afaik. it's done it on everything i've ever done [19:55] it does the debsign for that matter [19:55] which is why i have DEBSIGN_KEYID in ~/.devscripts [19:55] ok [19:55] but there i'll just debsign it myself i guess [19:56] i just don't understand why that's the case. it's super bizarre [19:57] should i NOT be uploading lubuntu-meta? is this actually generated? [19:57] Well, does the changelog check out? [19:57] Wait [19:57] Are you sponsoring this? [19:58] well i'm trying to figure it out [19:58] that was my intention but now things are going all weird [19:58] the changelog is your script [19:59] or some script [19:59] What is in the changelog? [20:00] this is the big update https://phab.lubuntu.me/rLUBUNTUMETAPACKAGING3cdce7d79da6891daa2f3879f8aada7b6e1fc177 [20:00] it seems messed up [20:01] but does that explain why it doesn't GPG sign? noooooooo [20:01] Nope, just dch -r [20:01] Then try [20:02] hah [20:02] it does explain it [20:02] *b*i*z*a*r*e* but cool [20:16] ~lubuntu-packaging wasn't subscribed to lubuntu-meta bugs *FACEPALM* [20:17] anyone want to go through and triage it would make my day https://bugs.launchpad.net/ubuntu/+source/lubuntu-meta/+bugs?orderby=-id&start=0 [20:23] some we can ix easy wxl [20:24] 'search: 14.04 OR Trusty OR Precise OR 12.04 OR {InterimReleases} | ACTION: Close. [20:24] :P [22:05] I'm working on Bug #1828663 -- looks to be a polkit problem [22:05] Bug 1828663 in lubuntu-meta (Ubuntu) "lubuntu 19.04 QT interfaces not properly working where more then one sudoer configured" [Undecided, In Progress] https://launchpad.net/bugs/1828663 [22:21] @TJ- [ I'm working on Bug #1828663 -- looks to be a polkit problem], damned politicians.... [22:24] Looks to be a segfault in lxqt-policykit [22:32] @HMollerCl you'll find network-manager-gnome is now in lubuntu-meta [22:33] goof [22:33] good [22:34] as soon as we have a new iso I will test it and if works change the defaut coneection-editor for nm-tray in nm-tray package. [22:35] might take a while given the apt-xapian-index issue [22:54] I came up with a 1-line fix for that :P [22:55] we didn't we have Tj- before? [22:56] TJ- [23:08] no but tj is a god send :) [23:10] +1 [23:12] (Photo, 1280x800) https://i.imgur.com/gQyRjhM.jpg Hello. This is not a Walpaper proposal, this is just a small experiment made from a change in the hummingbird image of the Lubuntu 1804 wallpaper. [23:13] i like the direction you're headed with that [23:13] me too [23:14] Hmm, makes me think about our SDDM background too. [23:15] I was just playing around 😊 [23:17] Here is some inspiration perhaps: https://github.com/Rokin05/SDDM-Themes/tree/master/.github/medias/screenshots/arc [23:17] With the humming bird image that @JyotiGomes has started on for a backdrop. [23:20] @JyotiGomes [], that looks really cool [23:21] I think that if later Lubuntu will have a logo of its own, replacing the lxqt logo (which is not specific to Lubuntu but rather to LXQt), it could well be this Lubuntu 1804 hummingbird. I do not know who created it, but it's fine. The hummingbird is a super fast, intelligent bird, active, effective and very beautiful 🙂 [23:26] @kc2bez [ Here is some inspiration perhaps: https://github.com/Rokin05/SDDM-Theme …], they work in lubuntu or are they only for plasma? [23:32] Here are the manual install instructions: https://github.com/Rokin05/SDDM-Themes#-others-environments--manual-install- [23:33] @kc2bez [ Here are the manual install instructions: https://github.com/Rokin05/SD …], Ok, thanks! [23:41] (Photo, 1280x800) https://i.imgur.com/NFShh6G.jpg Just more this version of color, I do not want to annoy you too much [23:49] not that one so much [23:50] With my Blind Society hat on, I'd ask you to consider how these images will be perceived by visually impaired people, who need strong contrasts and also have a better experience when certain colours are used [23:50] there's probably a design guideline for that somewhere.. [23:53] Re the policykit issue; looks like there may be two problems but I'm not familiar enough with the code bases to figure it out. 1) should a user already a member of "sudo" group even be prompted to select the user they wish to authenticate as? and 2) when they do something goes wrong and there's a SIGSEGV which looks like it may be caused in libpolkit-qt5 [23:54] I've updated the bug report but not sure I can take it much further [23:56] might have to dig upstream and see if there's anything sensible that shakes out of the trees [23:57] Already done that; there don't appear to be any relevant commits (in fact very few commits in recent times at all) [23:58] to be clear i'm referring to lxqt-policykit.. or maybe even the lxqt-config-users or whatever it is [23:58] right, I looked at that and libpolkit-qt5 [23:59] Looks like tsimonq2 is the committer on github for lxqt-policykit and the recent changes have been predominently translations [23:59] i know with lxqt there's often a lot of chatter that happens in issues, so sometimes i just dig in there for keywords and see if i can shake something out