[00:02] wow [00:02] all executors full [00:03] its gonna explode [00:04] swy-01 is scaleway? [00:06] 56 jobs all hitting us at once, this will be fun [00:06] yes, VC1S [00:07] the disk will probably run full soon. Though if this works well, I'll add a volume and just leave it running [00:16] well, that's that. Now i'm curious if they'll have new builds triggered eventually [00:16] there seemed nothing really wrong with the configuration as far as I could see [00:23] I'm upgrading my yy vbox [00:23] so far, only: [00:23] ./usr/lib/ [00:23] ./usr/lib/x86_64-linux-gnu/trojita/ [00:23] ./usr/lib/x86_64-linux-gnu/trojita/trojita_plugin_QtKeychainPasswordPlugin.so [00:23] ./usr/share/doc/ [00:23] ./usr/share/doc/trojita-plugin-qtkeychain/changelog.Debian.gz [00:23] oops [00:24] The following packages have unmet dependencies: [00:24] konversation : Depends: konversation-data (= 1.6-0ubuntu1) but 1.6.1-1ubuntu1 is installed [00:24] -f install seems to be fixing though [00:27] hm, you should only have to do that if something went wrong previously [00:28] and ok, I see a bunch of #3 builds for the ones that just got fixed, so it seems as if the triggers work now [00:28] now restarting it [00:28] yofel: this was restored from a snapshot; I don't think there were problems, but..... [02:13] o/ morw qtwebengine work for me [02:13] *more [02:15] thank you for that, tsimonq2 [02:19] lots of green [02:20] isn't it great, ahoneybun? [02:20] I love seeing it [02:20] it's never bad [02:20] more so when the green is money lol [02:21] I'd rather see green plants than green money, but then.... [02:21] privilege! [02:29] yea [03:56] lol then red [03:58] wait [03:58] then tehy get fixed? [04:07] anyone still on slack? [04:11] * valorie never got onto slack [04:19] your on there [04:20] yes, I have an account, but no team beyond Kubuntu [04:21] which -- we never really used it [04:22] I had used it for Mycroft [04:22] need to get back on that account [04:45] yofel: ok, publishing! thanks all! [04:47] let's refollow the autopkgtests later when the infra recovees from the shock of getting new frameworks & plasma [04:56] done https://ci-train.ubuntu.com/job/ubuntu-landing-041-2-publish/6/console [05:07] thanks Mirv [06:00] that's great Mirv :) [06:01] it's awesome that we're set up so early this time [06:01] at least hopefully [06:01] thanks Mirv [06:08] you're welcome :) [06:18] Mirv: what's going on? [06:20] tsimonq2: https://ci-train.ubuntu.com/job/ubuntu-landing-041-2-publish/6/console [06:22] plasma 5.7.2 and FW landing in the yakkety main archive [06:27] that's awesome [06:29] good morning [06:38] morning :) [06:45] yofel clivejo: I had done it locally, but left that cervisia KF5 port for Simon, as thought it could be a useful one for him to do [06:49] hey y'all [06:49] hi :D [06:58] chromebooks suck, I need to have a way to make my development environment more portable [06:58] (custom config files, keys, etc.) [06:59] it's annoying to have to set up on every computer I use, and it usually ends up being a terrible experience compared to my regular machine [07:00] so pretty much all I can do is read and talk, while not getting any work done, it's frustrating :( [07:01] but whatever, I'll be around if anyone wants to chat ;) [07:20] tsimonq2: you can get work done over ssh ;) [07:20] well, if you have access to the keys [07:20] yofel: I have no server to SSH to and my computer is off [07:21] tsimonq2: ah, we can fix former, latter... not [07:21] 03:30:56 /var/lib/jenkins/ci-tooling/kci/source_publisher.rb:44:in `wait': Upload was likely rejected, we have been waiting for well over 30 minutes! (RuntimeError) [07:21] hm... I think I need to increase that timeout [07:22] yofel: the thing is, I don't have my private keys on me and I need those to do a couple of important things [07:23] ahoneybun: because the builds are flaky as that, jenkins will try to build every job 4 times before concluding that it's broken [07:25] although I *can* do Git I think [07:26] yofel: there's some out of date symbols I can fix on the Update Excuses list for Qt 5.6.1 [07:29] would you rather I wait and do it properly or do you accept patches via pastebin? :D [07:31] tsimonq2: the others might accept pastbins ^^ - I'll probably going to be too busy today to look at anything before the evening [07:32] alright cool :) [07:32] and I can't type today o.O [07:35] you can sing ! :D [07:38] I'm siiiiiiiiiiiiiiinging in the sun, just siiiiiiiiiingin' in the sun, What a glooooorious feeling, and I'm ... fried again [07:39] ") [07:44] *SIGH* I can't set up apt-cacher-ng [07:44] it's gonna be a long day... [07:44] (can't start systemd processes because it's in a chroot) [07:46] * yofel wonders if systemd-run chroot ... would work [07:46] I usually work around it by throwing the cache into lxc if I need it seperated [07:46] yofel: can't run LXC on this thing either... [07:46] unless all you have is the chroot [07:46] ah hm :/ [07:47] I'm inside the chroot and that's all I can mess with [07:53] great, and I can't use sbuild either [07:53] I'm done with that [07:53] **/SIGH [07:54] *SIGH* [08:45] Good morning KUBUNTU [08:45] :) [08:45] GOOD MORNING JIM [08:45] :) [08:45] :D [08:45] HI Good Morning 😃 [08:46] It's Friday, and it is a lovely day [08:46] \o/ [08:46] Rick! :D [08:46] I'm actually watching the latest Kubuntu Podcast :D [08:47] Yay! [08:48] hm, with today's CI behavior, I have this feeling of watching a system that's quietly doing it's job, but somehow I don't trust that at all because that feels not normal at all ^^ [08:49] yofel: can we please rewrite in Python? Launchpad has a Python library :P [08:49] XD [08:50] as does ruby FWIW (self written by sitter ^^). But I wouldn't mind a language mix in there personally [08:51] at least I have done *some* python! [08:51] and Python at least doesn't look weird :P [08:51] Python is easy [08:52] hey Rick :) [08:52] I started on Fortran and Pascal! [08:52] Fortran OMG [08:52] LOL I hated that language xD [08:52] but now I started to appreciate its simplicity :) [08:53] jimarvan: don't you think we should port KCI to Python? :D [08:53] we should first build everything from scratch xD [08:53] hehehehehe [08:54] As an engineer I love "clean" solutions [08:54] that *was* a "clean" solution btw. ;) [08:54] i know i know [08:54] :D [08:54] jimarvan: I mean not Jenkins but the build scripts that do everything [08:54] still, learning ruby is less work then rewriting everything in python (which I don't personally consider superior to ruby) [08:55] gosh darnit :P [08:55] * tsimonq2 tried XD [08:55] lol [08:55] because at that point - we might as well do it the jenkins way and rewrite everything in java ^^ [08:55] at least then we would have a compiled language with static types [08:55] NOOO NOT JAVA D: [08:56] *that* I consider superior to both python and ruby [08:56] yofel: t'used to have a language mix. it's not a problem tbh. only reason it went away is because launchpadlib is obscenely slow and complicated to debug for the use case of KCI [08:56] Java is worse than Ruby :P [08:56] well, then... D, Rust, Go, whatever (C++ maybe not) [08:56] yofel: we should rewrite it all in Brainf**k XD [08:57] errrrrrrr.... piet? ^^ [08:57] huh? [08:57] just a simple easy to debug language would do :P [08:57] then we could print out the apps at least :P [09:00] so about QtWebEngine progress [09:00] 2p I would be OK with Python, and it is a very popular langauge. useful for newcomers [09:00] QtWebChannel builds fine [09:00] Most important though is how well documented the code is [09:00] I agree with Rick! yay Python! :D [09:00] and it's easy to document the code if you can actually read it XD [09:00] Sick_Rimmit: ruby code is self documenting >.> [09:01] * yofel runs [09:01] QtWebChannel just needs to pass some tests [09:01] we disabled them for the time being [09:01] Yep, I have no issues with Ruby either. I have written any, have read some, reckon I could get my head around it [09:01] but, it would be good to fix them [09:02] afair we still have a long ways to go on QtWebEngine, I don't remember if he actually got it to build or not [09:02] *it [09:02] wait nvm on that edit [09:02] but anyways [09:03] I say eta of ~ 1-1.5 weeks to get into Debian [09:03] it's cutting it close to FeatureFreeze, but hey, we'll have it :) [09:03] Sick_Rimmit: but documentation is probably the one thing the tooling lacks most. As I'm not very familiar with the ruby ecosystem, it took me hours just to make things *run* [09:04] yofel: which is why we should port to Python so it's easier to make docs :P [09:04] * tsimonq2 runs [09:04] * yofel sends a python after tsimonq2 [09:05] * tsimonq2 throws a ruby at yofel, hitting him, while he proceeds to trap the python and send it back at yofel [09:05] :P [09:05] touché [09:06] sounds like some off kilter DND/role play [09:06] huh? [09:06] FWIW, I don't really like python virtualenvs either, but they are a bit easier to use IMO than this rvm thing the tooling runs in on the builders [09:07] yofel: rbenv > rvm [09:07] sitter: why are you using rvm then? ^^ [09:07] rvm has the problem that it is basically black magic because it tries to do everything behind the scenes with little to no control [09:07] yofel: because that's what we originally used [09:08] don't change what isn't broken :P [09:08] our newer CI setups have rbenv [09:08] @yofel Perhaps that would make a good task for Ninja's to look at - Refactoring and documenting the tooling [09:09] or maybe I should do that to convince them to grant me ninja :P [09:09] (kidding, I agree with Rick) [09:09] ninjas is about packaging, not coding. If anything that counts towards regular membership. [09:10] Refactoring isn't really needed, the code works well, just figuring out how to use it is a pain [09:14] ...speaking of CI :D [09:15] ooops, that was me dist-upgrading swy-01 and not realizing that docker was on the upgrade list [09:21] screw it, I'm logging into c9.io [09:21] Ooo that looks interesting [09:22] yeah, I used to use it to access IRC at school [09:22] (for grades 7 and 8 I got to carry a Chromebook around) [09:22] but it's really cool [09:23] I even convinced a few teachers to let me code things for class projects, and I managed it all under Git [09:23] it was beautiful [09:23] but it's also nice for spinning up a cloud instance to just code on [09:24] I'll use it now to do what my development environment can't :) [09:24] (my CURRENT one, not the one at home) [09:26] let's see if C9 likes sbuild :D [09:27] Sick_Rimmit: One thing about C9 that I really like is that anyone with a GitHub account can collaborate on a terminal [09:27] (yes, multiple people can work on something) [09:27] I used that a lot with wxl [09:28] speeds aren't too bad either ;) [09:35] yofel: so how far after Qt 5.6.1 gets in the archive do we start porting to Qt 5.7? :D [09:46] 'Morning [09:53] hiho BluesKaj [09:54] tsimonq2: not this release. Next are finishing apps for yakkety (and maybe working on 16.08). And in parallel the plasma 5.7 xenial backport (including qt5.6.1, which is building in some landing for xenial) [09:57] oh alright [09:59] and i think 5.7 for Xenial is more important than Qt 5.7 in YY [09:59] *Plasma 5.7 [10:00] yeah I see [10:45] you know what, next onth i will try to create simple widget for plasma [10:46] so some basic stuff for the welcome/slideshow i was talking about some time ago [11:08] stderr: fatal: index file smaller than expected [11:08] brand new ways to fail [11:11] and all of those were on linode.. [11:14] maybe shoot a support email to linode asking why? [11:15] lets wait a bit more, and it's friday anyway [11:16] might be some automated feature kicking in, maybe thinks its a DDOS attack on KDE [11:16] could be. I tried setting some cpu limits on the container in an attempt to slow it down. But that for some reason seems to be ignored... [11:16] need to look closer at that [11:17] woooohoooo [11:17] OMG finally it worked [11:17] I am getting payed!!!! (celebration fireworks *****) [11:17] *paid [11:20] and again.. [11:21] yofel: don’t you find this maddening ? [11:22] I did until I added the auto-retry workaround, now I find it annoying [11:23] but there's more pressing issues that I want to solve, so I don't consider it that much of a problem [11:23] hmm :/ [11:23] such as? [11:25] tooling deployment doesn't work with multiple slaves and missing stable builds [11:25] 11:19:05 > /usr/bin/git -c core.askpass=true fetch --tags --progress git://anongit.kde.org/kmines +refs/heads/*:refs/remotes/upstream/* [11:25] yofel: you should use the git-monitor in tooling [11:26] sitter: speak english please [11:26] https://github.com/blue-systems/pangea-tooling/blob/master/git-monitor/git [11:26] implements a semaphore for git [11:27] run https://github.com/blue-systems/pangea-tooling/blob/master/git-monitor/git-monitor as a system service, and the git thingy will intercept network bound commands and semaphore lock them [11:27] prevents kde from rejecting connections [11:27] aaaah, thanks [11:32] sitter: is it KDE is rejecting them? [11:35] yup [11:35] we only allow 5 concurrent connections [11:37] ah [11:37] then does it ban for a certain period of time? [11:37] ^ thats useful to know :) [11:38] not that I know of [11:54] I wish once, to view IRC and see only green color [11:54] one day... one day... [11:55] adjust screen colors to render text in green :D [11:55] That collaboration feature looks excellent [11:56] I know jimarvan wanted to work with me on Sphero-Droid, that would certainly by doable in c9 [11:56] I'm taking a look further at that [11:56] Hmmm... I wonder if I can blag some sponsored accounts.... [11:58] jimarvan: the funny part is: 'SUCCESS' is almost never printed here, so if everything is fine, you would see one single message at midnight UTC and that's it ^^ [11:59] :D [12:03] :D [12:03] Rick [12:03] 16th of August I am on holiday in Greece [12:03] I will work on Sphero code and kubuntu while on the beach! :P [12:03] And everyone is welcome in Zakynthos ;) [12:29] @Sick_Rimmit, Playing with a toy is work?!?! [12:54] hi do you guys have plans to provide plasma 5.8 for kubuntu 16.04? [12:54] momentum: in the future - yes [12:54] ahh good to know [12:54] thanks for the info [13:03] see ya laters peeps :) [13:23] @Sick_Rimmit: Yeah, C9's collaboration is nice :D [13:56] I can confirm debian bug 832900 with the staging PPAs [13:56] Debian bug 832900 in plasma-workspace "plasma-workspace: Third-party systray icons get a black background" [Normal,Open] http://bugs.debian.org/832900 [14:14] tsimonq2: what are you getting blacked out icons for? [14:16] acheronuk: Chromium, but Franz is also messed up [14:16] Clifford: Outrageous [14:17] I just saw your reference to Sphero-Droid as a TOY!!! [14:17] Hmmmm. [14:17] I'll have you know that it is a very serious endeavour [14:17] heheheh [14:19] tsimonq2: don't use thsoe, or have a tray icon for them. just wondered, as shutter, thunderbird, hexchat and dropbox icons are just fine here [14:19] yeah [14:22] tsimonq2: clememtine mentioned in that bug report is also ok here. weirdness [14:22] yeah === flexiondotorg is now known as Wimpress === allee is now known as alleehol [16:02] Overlord II Now Available For Linux Gamers [16:21] * acheronuk peers meaningfully at marble [16:21] hmmm [16:22] off to bed very soon [16:23] acheronuk: I *think* it's symbols? [16:23] but I'm back at home! \o/ [16:23] o/ [16:26] it needs a new versioned libmarblewidget library package, and symbols regenerated for that I think [16:26] amongst other things [16:28] tsimonq2: you are either on a very odd time schedule, or not in the timezone I thought you were [20:20] hm.... [20:20] yofel: something going on with the git sycamore [20:21] /var/lib/jenkins/tooling3/git-monitor/lib/semaphore.rb:156:in `synchronize': undefined method `synchronize' for nil:NilClass (NoMethodError) (Git::GitExecuteError) [20:22] yeah, I'm looking at the log [20:25] hm, it's also calling the wrong git during the pull phase [20:26] sitter: around? [20:28] or do I have to set the PATH for the jenkins user on the slave ?!? [20:28] but that's not working on master either [20:36] I see code to special handle /var/lib/jenkins/tooling3/git, but that's not a thing [20:40] evening :) [20:41] hi Rik [20:44] acheronuk: you up to much? [20:44] symbols question (oh fuuuun!) [20:44] * clivejo runs and hides [20:44] if the libmarblewidget-qt5-23 library package needs to change to libmarblewidget-qt5-25 where libmarblewidget-qt5.so.23 has changed to libmarblewidget-qt5.so.25, then does that new library package inherit the old symbols file renamed as a starting point? [20:46] yes, and also renamed in the actual symbols file [20:46] there will be a header line that needs updated [20:46] at the top you mean? i.e. libmarblewidget-qt5.so.25 libmarblewidget-qt5-25 #MINVER# [20:47] yes, that! [20:47] yup [20:47] ok. I'm doing it right then! [20:47] while it's harmless to reuse the symbols file, usually the way is starting a new one [20:48] and keep the -bin dependencies in it (if any) [20:48] santa_: can you example the reason for that? [20:48] explain [20:48] -bin is one of the reasons why I *prefer* reusing the old file [20:48] and it doesn't matter which one you do in practice [20:49] yes [20:50] clivejo: the reason is that the symbols files are meant to generate dependencies in the packages linked against that library, so ... [20:50] re-using the old one made sense to me, and worked, but I then took a peek at what Neon had done with it, and they obviously started afresh [20:50] supose we have a library libfoo1 [20:50] which made me doubt [20:50] version 1.0.5 for example [20:50] then a version 2.0 arrives [20:50] with the soname bumped [20:51] so we have the libfoo2 binary package [20:51] the minimum version of libfoo is 2.0 [20:51] so here we have: [20:51] a) reusing the old symbols file updating it [20:52] b) starting a new one [20:52] yofel: presumably as it's changed 2 versions, #MISSING are a gimme for being removed? [20:52] acheronuk: yes, once the version has changed, all changes are OK [20:52] if we go for b) all the symbols will have a 2.0 version [20:52] cool [20:53] so if we have a package linked against it it will have a versioned dependency like this [20:53] libfoo2 >= 2.0 [20:53] this is right [20:53] now supose we go for b? [20:54] and we have the same package linked against a symbol listed in the symbols file we reused [20:55] clivejo: did you have anything in mind with the "you up to much?" question? [20:55] supose this symbol is listed as appearing in the 1.0.5 versipm [20:55] then we will get a dependency like this [20:55] libfoo2 >= 1.0.5 [20:56] since the minimum version of libfoo2 is 2.0 [20:56] the package manager will behave exactly the same when installing the package linked against libfoo2 [20:56] so how would you go about generating a new symbols file for it? [20:57] let me find that link... [20:58] clivejo: http://pkg-kde.alioth.debian.org/symbolfiles.html [20:58] using that commands AND [20:58] Id like to give it a try sometime [20:58] checking the -bin dependencies of the old symbols file [20:59] can you explain what you mean by checking the -bin deps? [20:59] like compare the two? [20:59] if in the header of the old symbols file there is a libfoo-bin dependency, you MUST add it to the new symbols file [20:59] and you have to do it manually [20:59] or right [21:00] oh [21:00] I tried starting a new one, and that is probably why it didn't like that one bit? [21:00] maybe you find easier the other approach which is also fine, as explained [21:01] also, in case you didn't already [21:02] https://www.netfort.gr.jp/~dancer/column/libpkg-guide/libpkg-guide.html [21:03] ↑ I think it's a MUST read [21:04] clivejo, acheronuk: any doubts? [21:05] ok. I had marble working once already in pbuilder, but then went off at a tangent on the new file thing. [21:06] I've had enough of that today, but I'll do a complete re-do tomorrow to get it in my head [21:06] and read the exciting docs [21:06] not really exciting, but like a medicine of bad taste imo [21:07] but hey, if you find them exciting, good for you, I'm not gonna demotivate you ;) [21:07] * clivejo giggles [21:07] to be clear, no I don't find that exciting! [21:07] but I've read worse [21:08] I sensed the sarcasm! [21:08] I propose reading the GNU make manual. Amazing collection of dustry dry topics to choke you to death by boredom really slowly [21:09] oh, and you would probably be the first one here that would manage reading the whole thing :P [21:09] I don't usually read between lines [21:09] should we give it to Simon to read? [21:09] in any case exciring or not the guide is something you have to read if you want to do a proper job with libraries [21:09] be like his initiation [21:10] anyway, I *think* I figured out how to use this git-monitor now [21:10] otherwise you might start to do bizarre things [21:10] like sending the newbie painter to go fetch you a tin of rainbow paint [21:10] but yeah, I agree about the lib packaging guide [21:11] I will bite the bullet [21:12] note that we also have that a particular way in kde packages to handle the abi breaks [21:12] https://www.youtube.com/watch?v=YbYIocYmSCI [21:13] that abi manager thing, but you can skip that for now [21:15] clivejo: mean but funny :D [21:19] stupid autotests [21:41] weird [21:41] spectacle has an epoch in KCI unstable [21:42] Published 2016-06-22 [21:43] * clivejo wont go there [21:45] delete it? [21:45] keep getting timeout on LP [21:45]