[00:24] hey clivejo and Clifford , when do you plan on applying to become a Kubuntu Developer? :P [00:24] Clifford ^ [00:53] Clifford and clivejo are the same [00:53] ik ahoneybun [00:53] :P [00:54] Since you know PyQt go fix the slideshow @tsimonq2 [01:02] Mm my screen locker is broken [01:03] mm [01:42] yofel: I see screenshots of Linux Mint 18 using a Qt UI for their installer [01:42] so they must have fixed it or are going it hit it too [02:03] wooooooah so many merge failures [02:52] ATTENTION NINJAS AND YOFEL! I'm rebuilding mgmt_merger because I believe that retries all the mergers. A lot of them are failing because of errors I'm fixing by rebuilding, so a global rebuild is just easier. Either way, I don't think it hurts. ;) [02:54] This is to make it clear that my above message was not a KCI output. [02:54] fun colors, tsimonq2 [02:56] valorie: well I do that when I want to get the attention of some people who are grepping logs [02:56] :P [02:57] valorie: see what I mean? :P ^ [03:13] oh that's great... [03:13] screen of red... [03:19] worked for the first few..... [03:20] valorie: tooling seems broken but nobody's touched it lately according to the Git history... [03:25] yofel: so the tooling is broken I think? ^ [03:32] ooh what's this? :P [03:32] woah [04:47] one more try on mgmt_merger just in case [04:51] wth... === sakrecoer_ is now known as sakrecoer [06:18] tsimonq2: WTF? [06:19] acheronuk: you talking about how broke KCI is or did I mess up? [06:19] KCI [06:19] yeah it's BROKEN [06:20] * acheronuk goes for coffee [06:43] not sure if that is the tooling or the slave [06:46] looks to me as if the linode node is the one where the errors are happening [06:49] lp is smart [06:49] found me among the contributors to qupzilla and added to the actual project [06:52] yofel: turn it off and on again so it can fix itself :P [06:57] acheronuk: maybe this is relevant? http://kci.pangea.pub/job/mgmt_docker_cleanup_amd64/ [06:57] that was disabled [06:57] I think [06:57] because it seems that went every two hours [06:58] or anm I wrong here? [06:58] well I'm wrong on the timing but still [07:01] you just triggered that, but failed on lindode? [07:01] yeah [07:01] same critical error [07:03] http://kci.pangea.pub/computer/ [07:04] Response time: Time out for last 2 try [07:04] so Linode is crapping out [07:04] not sure if symptom, cause, or neither [07:04] grrrrrreat [07:05] acheronuk: like I said, all yofel needs to do is turn it off and on again :P [07:05] or maybe that container at least [07:06] yeah [07:06] acheronuk: do you have access? [07:06] no, and I wouldn't mess with it if I had to be honest [07:07] clivejo, yofel: KCI is broken, flag [07:07] get it acheronuk? red flag? :D [07:08] * acheronuk groans [07:08] ? [07:13] tsimonq2: couldn't sitter help with that? [07:13] maybe [07:16] should it be linode rather than anything else, then I don't think he would have access [07:19] ok I'm up too late, I NEED to go to bed, o/ [07:24] o.O [07:25] acheronuk: I think everything needs a reset. There's errors on the slaves, but mgmt_merger only runs on master, I got a stacktrace trying to log in, ... [07:27] I was looking at those errors rather than that master job, but yes, none of it seems very happy! [07:30] lets try to just go with a CI reset [07:31] FWIW, clivejo has the same permissions as I do if this ever happens again and I'm gone [07:33] Noted. I saw he was on the ci-admins on LP [07:37] * clivejo is broken too [07:37] morning [07:41] * soee fixes clivejo with his magic toolbox [07:47] * clivejo thinks jenkins must run on windows [07:49] well, it's java so that should be fairly easy to set up ^^ [07:51] javax.servlet.ServletException: org.openid4java.discovery.DiscoveryException: 0x704: Failed to discover XRDS document from https://login.launchpad.net/+id/f6PQhez [07:51] again o.O [07:52] ok, let me reboot the whole thing and hope it'll boot back up [07:56] wow, I can barely use ssh on that box [08:13] the network connection feels horrible, but it seems to work overall for now [08:14] had any problems packaging kdevelop 5.0? I'm curious [08:19] kfunk: not sure if anyone has tried. blaze already had daily builds in his ppa, and Neon's JR said it was building om their CI last night. so I guess between those 2 most heavy lifting for debian packaging is handled? [08:27] yofel: did you get that updated package in my PPA? [08:27] not yet [08:28] looks like studio fixed theirs [08:32] yofel: what do I need to look at if I wanted to go for kubuntu-dev? [08:38] * acheronuk is pleased clivejo is seriously thinking about applying for dev :) [08:38] dont think Im ready [08:38] let me postpone that discussion to lunch, busy right now [08:39] but maybe I could study up on the areas Im weak on [08:39] yofel: no problem [08:51] ;o [08:54] that's better. I hope [08:55] oh dear [09:03] I *think* launchpad is having issues? [09:04] shock horror [09:12] it's too late, screw it, I'm staying up [09:15] clivejo: so you basically admitted you're ready for the hour of torture I keep hearing about? :D [09:15] Im *thinking* about it [09:15] you could read the old meeting logs, that's usually pretty informative [09:16] yofel: *I* struggled to find those [09:16] look at when people got added to ~kubuntu-dev, then look for IRC logs of here or #ubuntu-meeting of that day [09:20] Aaron Honeycutt 2016-08-01 [09:20] huh?!? [09:20] I didn't know he did dev stuff [09:20] https://launchpad.net/~kubuntu-dev/+members#active [09:21] ahoneybun: uhm.... did you press a wrong button? ^ [09:22] I was wondering.... [09:23] I will learn from the logs of clivejo's meeting ;D [09:24] me too :P [09:24] I was so generous to kick him out :P [09:25] acheronuk: I wonder if I catch up enough so your torture hour is before mine, so it's like a two hour thing, maybe we can secretly plot for that to happen :P [09:25] oh wow, even nixternal is still on the team [09:25] * yofel wonders what he's doing these days [09:25] lol. make scarlett the next newest [09:26] tsimonq2: you never know. one thing at a time for me [09:27] well, clive has the head start here as he's a kubuntu member already, you both need to get that done first ;) [09:28] acheronuk: again man, we should plan for the same meeting to torture everyone else ;) [09:28] yofel: exactly. one thing at a time. and I *know* there are several roadblocks in my knowledge besides. and that is just the things I know of! [09:29] well all I know is that symbols suck and I need to brush up on my C++ :P [09:29] tsimonq2: ditto [09:30] hey @Sick_Rimmit we need a ninja dojo session with clivejo and yofel :P [09:31] to teach us C++ and how to properly update symbols :P [09:31] *technically*, knowing c++ is not a requirement for ~kubuntu-dev. Knowing how symbol management works is though. So as long as you know the dpkg technicalities and basic linux shared object workings, you're good [09:32] * tsimonq2 rewrites everything in Python :P [09:32] yofel: well trying to work out where/why some of the akonadi symbols had gone missing, did show up I need to do some more C++ [09:33] not enough to be an uber expert coder, but a bit more targeted reading up on some things [09:34] ^ [09:34] I would at least like to know the Qt libraries [09:34] ok. bbl. stuff to do :/ [09:34] o/ acheronuk [09:35] clivejo: hm, that kubuntu-meta update is correct regarding the meta contents, but the changelog is rubbish o.O [09:35] but why.. [09:36] kubuntu-meta (1.342) yakkety; urgency=medium [09:36] * Refreshed dependencies [09:36] * Added packagekit to desktop-recommends [09:36] -- Adam Conrad Tue, 16 Aug 2016 18:01:58 -0600 [09:36] I don't see what's bad about that? [09:36] but I haven't read the diff... [09:36] no, I mean 1.343 which will remove okular [09:37] where's that at? [09:37] but it removed it from all archs, then wrote "[arm64 armhf ppc64el]" in the changelog... [09:37] germinate is weird [09:37] clive's ppa, and in the archive in a minute [09:38] yofel: why did you abort things? :O [09:39] because I wanted to reset things [09:39] ok [09:40] yofel: is debian-revision-should-not-be-zero a KCI error or Lintian? [09:40] sounds like lintian, someone uploaded a "3.0 (quilt)" package with a native version? [09:41] yofel: no that's on EVERY package in KCI [09:42] uh, o.O [09:42] look for yourself [09:43] tsimonq2: I don't see it? example please [09:44] yofel: https://launchpadlibrarian.net/280396680/buildlog_ubuntu-yakkety-amd64.cervisia_4%3A16.04.3+p16.10+git20160822.1857-0_BUILDING.txt.gz [09:44] E: cervisia source: debian-revision-should-not-be-zero 4:16.04.3+p16.10+git20160822.1857-0 [09:44] W: cervisia source: out-of-date-standards-version 3.9.6 (current is 3.9.8) [09:44] E: cervisia: debian-revision-should-not-be-zero 4:16.04.3+p16.10+git20160822.1857-0 [09:44] E: cervisia: package-must-activate-ldconfig-trigger usr/lib/x86_64-linux-gnu/libkdeinit5_cvsaskpass.so [09:44] E: cvsservice: debian-revision-should-not-be-zero 4:16.04.3+p16.10+git20160822.1857-0 [09:44] W: cvsservice: binary-without-manpage usr/bin/cvsservice5 [09:44] E: cvsservice: package-must-activate-ldconfig-trigger usr/lib/x86_64-linux-gnu/libkdeinit5_cvsservice.so [09:44] N: 5 tags overridden (5 warnings) [09:46] ah, kci ignores those I believe [09:46] but it's still very annoying in the build log :P [09:46] can it not upload -1 instead of -0? [09:46] well, you shouldn't read that for lintian warnings, you should read http://kci.pangea.pub/job/yakkety_unstable_cervisia/lastFailedBuild/parsed_console/ [09:47] WHAT [09:47] yofel: where did you find this?!?!?!? :O [09:47] acheronuk: did you know about this? ^ [09:47] that's amazing [09:47] go to a build -> click on parsed console output at the left [09:48] yofel: speaking of that, package-must-activate-ldconfig-trigger ? [09:48] wth? [09:48] false positive, question is what the proper fix is. [09:49] how did you tell it was a false positive? [09:49] s/did/could/ [09:50] those are kdeinit plugins, which don't need to be callable as regular libs as they're not actually linked to by anything. ldconfig ignores them by default as they're unversioned and throws that warning because it cannot determine what to do [09:51] so what do you do about it? [09:51] technically, those should not be in the public lib path, not sure why they are [09:51] IMO, just override the warning, it'll work fine [09:51] hm... we did have a dh_movelibkdeinit for kde4 in the past [09:52] something to think about, but I have to leave for now [09:52] tsimonq2: yes, I did know about that [09:56] tsimonq2: I assumed you did, but obviously not [09:56] :O [10:00] oh ffs... I want to get stuff done! [10:00] why do I have to wait for this stupid build queue... [10:03] I know. I know [10:03] finally! I reached the end of the dependency tree! [10:03] now I get to watch it slowly get to the code I just pushed :/ [10:04] 7 different jobs stacked :/ [10:05] LP taking forever to publish really does NOT help [10:05] and of course, LP is the cause of my pain and suffering once again [10:06] acheronuk: I asked on #launchpad recently about this. Someone's working on better parallel publisher support, but there's no artificial buffer. Launchpad literally has that queue size... [10:07] yes, cjwatson basically said the same to me [10:07] yeah that's the guy I talked to [10:07] it's a pita [10:08] * acheronuk wishes for another 4/5 nodes on linode so can have our own build farm [10:08] YES! [10:08] :D [10:09] acheronuk: if in another couple months we don't see progress, we should get our own package building farm, that would really help [10:10] start saving your pocket money, or find us some sponsors with pockets big enough [10:12] hehehehe :) [10:14] Neon seems a fair bit slower on actual builds than LP builders, but they don't have these wait times AFAIK [10:20] unfortunately I did not managed to get qupzilla working with acheronuk's qtwebengine :( [10:21] even after rebuild [10:22] blaze: something wrong with it? not my actual packaging on that [10:23] acheronuk: it seems so [10:25] you don't need particulary big pockets for a build server, a reasonable one would be e.g. a scaleway C2L node, which is like 24€/month [10:25] blaze: what is wrong? [10:26] the question is more the technical side. Go with neon approach and have no way to manually upload something, or setup debile or wanna-build as a build scheduler [10:26] yofel: what is the most efficient way to test KCI against a new Qt version? [10:26] acheronuk: it looks like QWebView is not starting, idk why, no error messages [10:26] upload a new Qt to the PPA [10:27] or go the ugly way and add another PPA as a dep [10:27] tsimonq2: did you build any more of Qt 5.7? [10:28] blaze: just trying to build here. how to test that? [10:28] acheronuk: no, I never finished and tbh I got distracted [10:30] yofel: would you be able to create a ninja PPA for that? we would all upload, then when all uploaded, add the PPA as a dep and work to solve all the errors that arise just by the new Qt version. Then, when that's done, someone copies those packages over. Or, would you rather I get all the Qt packages good to go and just copy over with no testing? [10:30] your choice :) [10:30] blaze: ooh. flash flash and nothing much happens. I see [10:30] acheronuk: exactly [10:31] tsimonq2: you did QtWebEngine. any ideas? [10:31] nope [10:31] acheronuk: I worked on the debian dir not the rest of the code [10:32] :P [10:33] I know, just you might know if there have been any fixes discussed about it, either in upstream changes they might need to add, or in the packaging [10:34] * tsimonq2 shrugs [10:35] I thought unlikely, but had to ask [10:35] :) [10:36] blaze: If I can summon up the motivation, I'll maybe try compiling it in a Neon VM with whatever c=version they have [10:37] bit at the moment I clearly can't even type :P ^^^^^ [10:37] *but [10:37] FFS! [10:39] :) [11:10] /me wonders why yofel kicked me our of devs [11:10] because you are not one? [11:10] that is not a free-for-all team [11:11] yofel: so...what do I do about Qt 5.7? [11:12] this is true [11:13] tsimonq2: preferably take what's in debian experimental and try to build that in yakkety, otherwise try to use what neon has [11:14] ahoneybun: that's why even the council is only an indirect owner of the team, so that the coucil members themselves don't count as members [11:15] yofel: but I mean where? [11:15] yofel: what should I do? [11:15] tsimonq2: put it in a ppa of yours, and it everything looks good sync it to the CI [11:15] *if [11:15] ok [11:18] yofel: because this will likely happen within an hour or two, is it worth doing a global rebuild to pick up the changes or would you rather wait for the nightly? [11:19] tsimonq2: what will happen in an hr or so? [11:19] I really doubt that you'll get Qt done within an hour or two, but feel free to do a global rebuild [11:19] ok [11:19] acheronuk: Qt 5.7 sync from experimental [11:20] tsimonq2: not a hope in hell. the bootstrap and then rebuilds takes best part of a day, if LP is not slow [11:20] * ahoneybun dislikes the screen locker being broke [11:20] heh [11:20] ahoneybun: where? [11:21] that should like, never happen [11:21] acheronuk: worth trying right? :P [11:21] oh yakkety [11:21] great [11:21] no clue why [11:21] tsimonq2: do it, but be prepared for much waiting [11:21] ok acheronuk :) [11:21] can I go on vacation until academy and not care about broken software? yes? [11:22] my screenlocker on YY is just fine [11:22] idk yofel :P [11:22] tells me to go to ctrl + alt+ f2 and use loginctl unlock-sessions [11:23] right, that means it crashed [11:24] ahoneybun: I had that on and off switching session with 5.6.5 [11:24] it just happend with its been 5 mins so it locks up [11:24] 5.7.x has been ok so far.... [11:25] also I have a okular update coming [11:25] okular is to just try to stop a build problem on the iso [11:26] so it's safe? [11:27] it wanting the old kde4 kactivities messed up the iso build, so yes, 'should' be safe [11:27] mm [11:28] yofel: I've turned off locking for now [11:29] okular will be from the PPA [11:29] there was no archive upload for okular [11:29] I don't tend to bother with locking nowadays, but it's a hard habit to break [11:29] well by default it does [11:30] oh, yes it does [11:30] Howdy all [11:31] ahoneybun: I've never really left that to the system. I used to work somewhere where leaving your workstation unlocked was a serious disciplinary offence. [11:31] o/ BluesKaj [11:33] * yofel also needs a lock screen at work - it's not considered an offense, but there are things like unlocked ssh keys etc. [11:33] hey tsimonq2 [11:39] ahoneybun: what version of plasma have you installed? [12:32] slhk: o/ [12:33] would you mind joining #kde-neon ? [13:06] yofel, clivejo, acheronuk: I'll get some spare testing done but otherwise Beta 1 needs testing! [13:07] right, go ahead and get lubuntu done at least [13:07] that's my #1 priority [13:07] don't you guys have a !testers thing? I thought I saw that once here... [13:07] * tsimonq2 is scred to ping all sorts of people :P [13:07] we do, but we have no images [13:19] yofel: I sort of forgot, what did kubuntu_stable do and why is it disabled? [13:21] meh, our seed is still busted, thanks to kde-runtime [13:21] tsimonq2: packaging for the kde bugfix branches (for apps and plasma) [13:22] but the CI tooling has no way to determine what the correct upstream bugfix branch name is right now [13:22] so disabled until that's fixed [13:22] yofel: ...isn't that a simple fix? [13:22] not if you have to rewrite code in a couple places :P [13:23] because the current logic relies on a branch namespace that we don't have on launchpad [13:23] I would help but I suck at Ruby [13:24] just like Ruby :P [13:24] everyone here does, that's the problem -.- [13:25] maybe I should really start and rewrite that in java [13:29] yofel: I'll do most of the work if it's in Python :P [13:29] I LOATH Java [13:30] well, then propose another language that has a compiler, which disqualifies python [13:30] we could go with rust? seems to be fancy right now [13:30] I don't know Rust [13:31] yofel: why does it have to use a compiler? [13:31] I only went for java because of jenkins anyway [13:31] compilers are slow :/ [13:31] tsimonq2: so I can actually rely on stuff at least executing without me having to write an extensive testsuite? [13:31] yofel: correct, if I'm understanding you correctly [13:32] I can help you write unit tests if you want, it will certainly help with making sure it's always stable :P [13:33] we have those in ruby, just unmaintained. Which is why I see little point in replacing one script language with another one [13:33] Python is very easy for beginners [13:33] as is java [13:33] no it isn't [13:34] I'm a beginner in Java and it's harder than Python :P [13:34] I know very little [13:34] but I enjoy Python because it's very flexible [13:34] yofel: https://twitter.com/notnownikki/status/767806785637982208 [13:35] right, it's a matter of opinion. I personally know java better than python, and I don't really like the way python does OO markups [13:35] https://spyurk.am/uploads/images/scaled_full_59d48025d37ced7b8028.gif [13:35] It depends on with which other languages you are familiar. If you have experience in other C-like languages, Java might be easier. If all your experience is in C-like languages, the syntax of Python can be extremely difficult to understand. [13:35] lets go with... D? [13:37] * yofel throws special characters at soee [13:37] don't get me started on Perl :P [13:38] In my opinion, Perl is a write-only language. :) [13:38] it is [13:38] :D [13:39] congrats, we have 270 something items in the build queue and like 3 actually running :/ [13:50] if yakkety_unstable_ktextwidgets keeps going for another half an hour, I will kill it with a sword [13:50] that change that I pushed HOURS ago? [13:50] not built yet :/ [13:50] in fact, all of those are really tempting to just make DIE [13:51] well specifically yakkety_unstable_ktextwidgets yakkety_unstable_kbookmarks yakkety_unstable_kio [13:51] then cervisia will FINALLY build [13:52] blame the really deep dependency tree [13:52] it's disgusting and needs to die. :P [13:53] just be happy that you're working on cervisia, not kdepim :P [13:53] or isn't there a way to prioritize this stuff?!? [13:53] yeah agreed :P [13:53] yofel: dude, can I just kill those three, no code changes, and green [13:53] PLEASE [13:53] it is prioritized by dependency levels, leaves build last [13:53] :( [13:55] well, feel free to, but fix whatever may break afterwards by building with older versions of dependencies [13:55] I will [13:55] I promise [13:56] * yofel went and crippled kactivities 4 [13:56] why does that still have to be a thing.. [13:57] FINALLY [13:57] it's building! [14:06] yofel: please tell me there's a tool to craft changelog files from scratch... [14:07] (as in debian/changelog) [14:07] yofel: I packaged QtSpeech and I think it's done except for that [14:08] depends on what you're trying to do, the format is simple enough to write just by hand. Usually I think people start that command that creates the whole packaging template [14:08] well I know the syntax [14:09] I'm just wondering if theres' an easy way to search the program to generate that file [14:09] from... what? [14:10] if you're doing a scratch upload, people usually just write "Intial upload (Closes: #whatever)/(LP: #whatever)" and are done [14:10] OH SHOOT I mean copyright files [14:10] sorry [14:10] any way to generate copyright files programatically? [14:12] oh that, not... really... there is licensecheck, and there's helper tools lying around somewhere, but as there is no standard for writing copyright headers, that's a bit tricky [14:12] maybe ask maxy what's he's using [14:17] clivejo 5.7.2 last time I looked [14:34] you guys should learn... (dramatic pause) Haskell! [14:34] EEW! [14:35] uhm, no. just no. [14:35] lol [14:36] I can't agree more with yofel [14:38] yofel: I see what you did there you mistook a . for an ! and didn't turn caps on before typing ;) [14:38] true, I was reading haskell, thinking ocaml, and my brain crashed [14:39] well that's 'cause yofel's brain runs on Java [14:39] and Java is complicated and hard :P [14:39] * tsimonq2 runs [14:39] * yofel likes the idea that his brain runs on coffee [14:40] hehehehehehehe, you run on coffee AND the convoluted, complicated programming language. EVERYONE WINS! :D [14:40] we could use scala :P [14:41] wth? [14:41] what's that? [14:41] another language that uses the JVM [14:41] EEW JVM is slow and terrible [14:44] we already use the jvm for jenkins, so the difference would be negligible [14:44] yofel: how good is your C? [14:44] I'll agree that jenkins is slow and terrible though :P [14:44] +1 [14:45] tolerable, but do we need to use a language that was only written because the machine it was made for couldn't handle a compiler for a language with more features? [14:45] I know, for something unrelated to this [14:46] well, depends on what.. [14:47] yofel: see #ubuntu-release [14:47] oh netcfg, haven't looked at that in a long time ^^ [14:47] pplleeaassee?????? [14:48] not while I'm at work, I need to concentrate when reading C [14:48] fine fine [14:48] yofel: if nobody ends up doing it, when do you get off of work? :D [14:49] in ~2h, then dinner, then driving home for an hour, then take care of some thing, look at kactivities again, play some WoW, and then I can look at netcfg if it's not midnight yet [14:50] ok lol [14:50] yofel: I have to trick infinity into doing it :P [14:51] I think he already got tricked :P [14:51] hmmm we'll see :P === santa is now known as Guest28643 [15:07] yofel: it would be more useful if you pined the actual person who's supposed to be doing that for the release team :P [15:07]