[09:41] clivejo: needs symbols sort I assume http://kci.pangea.pub/job/yakkety_unstable_kcalcore/ [09:41] *sorting [10:30] acheron_uk: know how to patch symbols? [10:32] I tried and got http://paste.ubuntu.com/17086159/ [10:35] clivejo: with the actual diff being http://paste.ubuntu.com/17086184/ [10:35] assume the +#MISSING needs resolving [10:36] Usually if it is a private symbol, we can ignore it [10:37] theres a link somewhere that we can check symbols [10:37] yeah, I got that, but went round in circles trying to check [10:38] a link? the API ref, or something more 'friendly'? [10:38] did you feed in both logs? [10:38] yes [10:50] acheron_uk: there looks to have been a SO bump recently - https://git.launchpad.net/~kubuntu-packagers/kubuntu-packaging/+git/kcalcore/commit/?h=kubuntu_unstable&id=4e4b61d7d16ada4126e145ca0e2091d8a22a5b30 [10:51] so that might be ok [10:51] remove it anyway and just make a note in the changelog [10:52] fair enough. :) [10:54] yofel: can you shed more light on this? ^ ^ [10:57] acheron_uk: are you subbed to the KDE Release Team ML? [10:58] I did sub to a few - have to check [10:58] Did you see this ? https://mail.kde.org/pipermail/release-team/2016-June/009483.html [10:59] you can sub here - https://mail.kde.org/mailman/listinfo/release-team [11:00] Yes I got that email, but admit I hadn't read it [11:01] LOL [11:01] thats my problem, too many emails! [11:01] I get so many emails on various things, that some filters are set to mark things as auto-read [11:01] i.e. they are there for ref if I need them [11:02] better turn off that filer [11:02] *filter [11:02] yeah, soee_ is supposed to announce this stuff! [11:04] for the note on the MISSING in the changlog, do I need to list the whole symbolname? [11:05] that seems cumbersome and untidy, but complete [11:05] no, just that you think its safe to remove due to the recent SO bump [11:06] is that from the CI build log? [11:06] yofel: yes [11:07] on a first guess: I would say "bad" - because the so bump happened for 16.04 already, which has already been released so doesn't cover this [11:07] but I didn't look at the actual symbol yet [11:08] it looks to be replacing and existing [11:08] also, huh? https://git.launchpad.net/~kubuntu-packagers/kubuntu-packaging/+git/kcalcore/commit/?h=kubuntu_unstable&id=ff4837985fae4f63a45f326efc48c53c21044c3e [11:10] * acheron_uk goes to make coffee [11:10] in that case which symbols do we take [11:11] we had the KCI version in kubuntu_unstable and 16.04.1 in yakkety? [11:14] oh, that was a merge? Then fine [11:14] yeah "Merge branch 'kubuntu_stable' into kubuntu_unstable" [11:14] I was just irritated as such things normally shouldn't happen [11:15] our work flow hasnt been very fluid recently [11:16] when apps got staged it didnt seem to pull in the unstable changes [11:17] I dont think Scarlett used the script [11:18] actually [11:18] https://git.launchpad.net/~kubuntu-packagers/kubuntu-packaging/+git/kcalcore/commit/?h=kubuntu_unstable&id=4e4b61d7d16ada4126e145ca0e2091d8a22a5b30 [11:18] has a misworded changelog [11:19] that was a lib version bump, *NOT* an SOVERSION bump [11:19] otherwise the bottom line would've changed [11:20] So to me, this looks like an ABI break introduced by 196f5d638be7ac4b0b085c5e09e41d6ef3927c4b [11:20] - _ZN8KCalCore10ICalFormat13readIncidenceERK10QByteArrayPNS_13ICalTimeZonesE@Base 15.07.90 [11:20] + _ZN8KCalCore10ICalFormat13readIncidenceERK10QByteArray@Base 4:16.04.1 [11:20] you might want to ask volker what his intentions were here [11:22] yes, he removed a function parameter, that's an ABI break as that function is in KCALCORE_EXPORT ICalFormat -> public: Incidence::Ptr readIncidence(const QByteArray &string); [11:22] Ive been getting crashes in korganizer I wonder are these related [11:22] you can't get any more public than that [11:22] in KCalCore::ICalTimeZoneData::ICalTimeZoneData [11:22] Howdy folks [11:22] only if the actual crash says "missing symbol in ..." [11:23] hey [11:23] hi BluesKaj [11:23] hi acheron_uk [11:23] clivejo: https://quickgit.kde.org/?p=kcalcore.git&a=commitdiff&h=196f5d638be7ac4b0b085c5e09e41d6ef3927c4b [11:24] Remove ICalTimeZones from public API here. [11:24] It's supposed to be removed as part of John's work to kill KDateTime and [11:24] KTimeZone, so hide it inside KCalCore here where it's easier to refactor. [11:25] sounds sensible, but requires an so version change [11:25] so maybe politely ask him whether he's aware of that [11:26] sometimes people just don't realize it [11:26] politely? [11:27] well, something like "hey, I think your commit foo to bar breaks the ABI, did you forget to change the so version or is that intenional?" [11:28] the plasma-nm dev I talked to also didn't realize that back with his wimax refactoring [11:29] that was back in April? [11:30] yeah, doesn't look to be part of the 16.04 branch though, so at least it's master-only [11:31] ah right, they branched end of march [11:31] so maybe just ping him that he should please not forget to bump the so version before 16.08 [11:31] or rather just bump it right now [11:35] clivejo: we break api/abi yes [11:36] is this a which comes first thing, the chicken or the egg? [11:38] depends on what the chicken is ;P [11:38] there's no rule for that really, hence why I said "polite reminder" because people tend to forget about it. [11:38] they seem to know about it [11:38] I would really recommend not removign the MISSING tag until the version changes [11:38] just so that WE don't forget [11:40] so in this case you would leave the #MISSING tag in the symbols file as a reminder? [11:40] yep. I hope lintian complains about that... haven't tried that in a while [11:40] Im sure it will [11:41] it usually finds numerious ways to complain about stuff! [11:42] acheron_uk: did you catch all that? [11:42] I'm here taking mental notes [11:42] my mental notes are broken [11:43] * clivejo has to write stuff down [11:44] so basically do the patch and accept that the #MISSING part stays for now? [11:44] right [11:45] yup, and make a note in the changelog that you expect a so bump [11:46] breaks stuff for people that try to actually use the CI packages, but the workaround for that is more work than it's worth [11:48] so something like? http://paste.ubuntu.com/17087001/ [11:49] that's fine [11:49] phew! [11:51] any way to get these hook/triggers fixed? [11:51] which ones? [11:51] LP git ==> KCI rebuild [11:52] interesting: " The wikis are locked down to only two teams: ~ubuntu-members and a Canonical team." - That would explain some things [11:52] * clivejo thinks we should get our own wiki [11:53] I cant even log into it [11:53] well, we use the KDE one for moderately updated stuff or docs page preperation. Notes for fast editing, and docs for static stuff [11:53] but yeah, the ubuntu wiki is a mess [11:54] surely a mediawiki install auth users against LP would be lot simpler [11:54] but I'm fairly sure that nobody here wants to maintain a wiki. Although we could restrict the login to LP teams as well [11:55] here we go again "Launchpad is temporarily unavailable" [11:56] set up a wiki on local machine here ages ago for personal notes, and found it was 10 x hassle than it was actually worth [11:57] for personal stuff I agree [11:57] but for public information and multiple editors, they are great [11:57] oh I know, and agree [11:58] was replying more to the "I'm fairly sure that nobody here wants to maintain a wiki" part [11:58] unless you get an edit war [11:59] yofel: mind if I stage FW5.22 ? [12:00] sorry 5.23 [12:03] go ahead [12:04] can I remove all the packages in staging-frameworks? [12:04] Ill go have lunch and start it running this afternoon [12:07] var-lib-snapd-lib-gl.mount is a disabled or a static unit, not starting it ??? [12:08] dpkg: error processing archive /var/cache/apt/archives/nvidia-opencl-icd-340_340.96-0ubuntu6_amd64.deb (--unpack): [12:08] there is no script in the new version of the package - giving up [12:10] ok, think it's the kernel version I'm using, it's non default [12:11] probably not. they reverted a nvidia change as it caused just such a bug [12:12] problem is it can need a bit of a hack to get the fixed packages installed [12:12] LP: 1589006 [12:12] Launchpad bug 1589006 in snapd (Ubuntu) "Failed unmounting Mount unit for nvidia support in snappy" [Critical,In progress] https://launchpad.net/bugs/1589006 [12:12] didn;t know my nvidia-340 driver needed fixing [12:13] https://launchpad.net/ubuntu/+source/nvidia-graphics-drivers-340/340.96-0ubuntu6 [12:14] don't tell me kubuntu is going to adopt snappy [12:15] maybe i should get rid of theubuntu-graphics ppa , i remembered that it's still enabled [12:17] that made no difference [12:22] there are workarounds in that bug report [12:24] ok there's nothing broken here in terms of multimedia [12:25] think I'll just leave it for now [12:39] well, we're looking at snappy. Although I'm not particulary impressed from what I've seen so far [13:09] hey guys, is Kubuntu 16.04 using Muon or Muon Discover? [13:09] both [13:09] are they separate apps? [13:09] it was orginally called Muon [13:10] but the packager manager is Muon Package Manager [13:10] and Discover is the software centre [13:11] ok, I think I understand now, thanks clivejo [13:11] the package manager was kinda left behind and non-maintain [13:11] clivejo: (reading the backlog) FWIW the community team is pushing for a switch to MediaWiki for wiki.ubuntu.com [13:12] but a few people have stepped up to maintain it [13:12] mhall119: any progress on that? [13:13] clivejo: we need to make sure it has a juju charm and mojo spec for IS to easily deploy and scale it [13:13] then the big effort will be in moving/converting all of the content from moin to mediawiki [13:13] since they use different markup [13:13] there are some scripts out there that claim to do this, but they'll need some heavy testing [13:14] Id be against the automatic import of all the moin data [13:14] yofel: if you have any questions about snappy I can try and answer them, otherwise the experts in #snappy are very helpful [13:14] theres large parts of it are extremely out of data [13:14] clivejo: that was also discussed, try and fing the most actively edited & viewed pages, and only copy them over [13:15] most edited we can get, but I don't know if we have data on most viewed [13:15] personally Id ask the community to move it over [13:15] manually? [13:15] and update everything as they go [13:17] I don't think the docs team would go for that [13:19] mhall119: thanks for the offer, but I talked with didrocks lately and he pointed me to the WIP docs, I just couldn't look into them in depth yet. [13:20] but stuff like dependency definition for snappy seemed... very maintainer work intense. But maybe I just didn't properly get how it works [13:27] yofel: not sure what you mean by "dependency definition" [13:28] all I've had to do so far is list package names from the archives that needed to be included [13:37] mhall119: ok, what I saw in the example IIRC was "stage-packages: libqt5gui5" - which is nonsense IMO as ldd can tell you that. And with the several dozen deps we often have thanks to the split frameworks etc. maintaining that doesn't sound fun [13:38] yofel: stage-packages is the equivalent of Depends in debian/control [13:39] except that *that* is auto-generated by dpkg-shlibdeps [13:39] it pulls dependencies from their package in the archive, and unpacks them into the snap's filesystem [13:39] ok, I see what you mean [13:39] yeah, for krita for example I had to check the build package's Depends, not the source of debian/control [13:40] which did require a krita 3.0 deb package already exist [13:40] I mean, I can understand that the system might have a hard time figuring out what ldd components are part of the foundation, and what is supposed to be included into the snap [13:41] but I would essentially need to write a script, that gets a whitelist of all binaries I know of that we might want to have included, run that over the compiled output, filter ldd, and then have that list all wanted packages in the field [13:41] would do the job, but sound somewhat clumsy [13:42] is that what dpkg-shlibdeps does? [13:44] to some extent yes, is runs ldd over all binaries, then asks dpkg in what packages said dependencies are (plus other stuff, but that's essentially it) [13:44] that can probably be done in snapcraft too, you can raise a bug on lp:snapcraft for it to do it [13:45] then a variable used in stage-packages like in debian/control that will put them in [13:45] at build time [13:46] right, I would hope someone's already looking at that, I simply didn't check yet [13:46] is kubuntu adopting snap or snappy or whatever it's called? [13:47] maybe someone is, but there's a lot of work going into snapcraft right now, so having a bug to track it would be good [13:47] BluesKaj: Kubuntu has the tools already in 16.04's archive [13:48] BluesKaj: additionally maybe, as a potential replacement for the backports eventually. If things turn out to work better than they do right now [13:48] thanks mhall119, yofel [13:48] it's still early days, but we're using KDE apps as real world use cases to work through the limitations and bugs in snappy [13:49] heh [14:57] clivejo: will KCI pick up the the new version numbers 5.23 for frameworks on it's own somewhere, or will someone have to update each component? [14:57] acheron_uk: Thats why I was wanting to do the staging [14:57] but Im having trouble with the script [14:58] That was what I hoped [15:01] http://paste.ubuntu.com/17090243/ [15:11] clivejo: just tried and got the same [15:12] not that I really know what I'm doing with that [15:39] clivejo: missing comma in https://git.launchpad.net/~kubuntu-packagers/+git/kubuntu-automation/tree/upstream-names.json [15:39] line 30 after libkleo [15:40] "libkleo" I mean [15:52] ah thanks [15:56] had to be something simple I thought, but my python is not great so slightly amazed I worked that out [16:14] few big downloads [16:15] probably just as well it rejected my ssh key when I tested then [16:16] just telling you what I been telling you :P [16:16] what? [16:17] become a kubuntu member :P [16:17] bit early to do that [16:17] why? [16:18] don't you need 2 months sustained contributions or something? [16:18] yeah, but you can start getting your profile and set a date for meeting [16:19] good point [16:19] you have a good understand [16:19] probably more than me! [16:20] don't know about that, but I generally learn quickly [16:21] ok staging script completed [16:21] 3 problem packages [16:23] did you get a KDE ID? [16:23] yes. but not a developer/contributor account yet [16:24] dont think you need that for depot [16:24] I thought you did, but could be wrong... [16:26] I don't see any option to upload a ssh key, except in the dev account application https://identity.kde.org/index.php?r=developerApplication [16:27] as and far as I can tell, depot tells you to get lost without one registered [16:28] it's 6 months actually - no exceptions, except that any kind of kubuntu related activity is valid [16:28] fair enough [16:29] https://community.kde.org/Kubuntu/Membership [16:29] If you think your contributions have been significant and sustained, you can try to become a member. Generally, a minimum of two months of visible, significant activity will be required. [16:30] yofel: so that 2 months is wrong? ^^^ [16:31] Im not sure, but theres lots of stuff needs doing :) [16:32] I only decided to go for it after valorie bribed me! [16:32] acheron_uk: yes, I guess that page is a bit outdated and needs fixing. The official guideline is https://wiki.ubuntu.com/Membership/NewMember - "it is rare for applications to be accepted from people contributing for less than 6 months" [16:32] and so I could push buttons on KCI [16:33] yep. I'm not overly bothered. would be nice, but it's not the reason I'm here [16:33] the borderline is a bit fuzzy, as you can never give a fixed starting point for "contributing", but we once got told "please don't do that" when we accepted someone after 2 months [16:35] I can understand that [16:36] they want people who are going to carry on with it it, rather someone putting in a short amount of effort to gain a 'badge' [16:37] exactly. You get a bunch of benefits, as well as a bunch of additional permissions with the membership, so it shouldn't be given out lightly. Makes it more valuable to have too ;) [16:38] Indeed. it has to be meaningful [16:50] acheron_uk: but you need to start on that path [16:53] think I already have [16:53] indeed :) [16:54] but yes, if going to end up there then it's best to have it in mind [16:54] get ducks in a row as they say [16:55] keep posting up git diff's [16:56] if they make sense I have no problem pushing them for you :) [16:56] I shall, and hopefully at a better pace once I have a better idea how to prioritise [16:58] i.e. things that are blocking the building of several others [16:59] do you have a change that you want applied right now? I would like to try having Launchpad be of use here ;) [16:59] acheron_uk: so Ive run the staging script [16:59] at the moment it feels like taking random potshots, which it not bad per se, but I would hope can do it with more reason [16:59] and I have a bunch of files I have to upload to the PPA [16:59] yofel: not this sec [16:59] Im about to go the git push [17:00] so I need to pause KCI [17:00] clivejo: yes. imagine that would be wise [17:00] acheron_uk: how much do you know about git? [17:01] as in: do you know what a pull request is? [17:01] yofel: very much beginner, but I do learn well [17:03] yofel: basically a request to include your changes? [17:04] yes, the better git hosting solutions like github and gitlab make heavy use of them for reviews, and LP can do them as well by now. [17:04] most things I done with git are just messing around with a local copy for my own purposes [17:04] That's I think a better way to do this than sending diffs around ;) [17:04] Now I need to figure out how to properly use them on LP though... [17:05] can you tell me how as well [17:05] If you can, and I am able to submit them, then yes sounds better [17:05] this move to LP git is all new to me [17:06] * clivejo opens up a new kate document [17:06] * clivejo kicks KCI [17:07] its too quiet [17:07] its like a bad child, when it goes quiet its time to worry [17:08] isn't it usually done with the daily builds around this time? [17:08] noooo [17:08] only gets done about an hour or two before it starts over again! [17:09] about 22:00-23:00 my time [17:09] So, I tried making this: https://code.launchpad.net/~yofel/kubuntu-packaging/+git/kcalcore/+merge/296702 by cloning kcalcore, committing, running "git push lp:~yofel/kubuntu-packaging/+git/kcalcore kubuntu_unstable", going to the repository branch site for the repo I just pushed to and opening a pull request [17:09] the annoying part is that you have to manually fill in the merge target :/ [17:09] github can guess that.. [17:10] but I guess it would do the job [17:11] now I'm off for dinner though. bbl [17:11] * acheron_uk get ready to try that [17:12] * clivejo wanders off for dinner too [17:12] well I got an email for your test [17:13] at least the nvidia upgrade bug isn't preventing other upgrades from working [17:14] so far [17:26] LP: is not responding to any clone I try to do, and some error pages on the website. [17:26] so I think time for refreshment here as well [17:46] Large whiskey! [17:53] tempting [18:14] * genii sprinkles some in his next coffee [18:17] clivejo : yofel think that worked [18:28] https://code.launchpad.net/~acheron/kubuntu-packaging/+git/kcalcore/+merge/296709 [18:56] so now that is rejected or accpeted, whet do I do with that repo/branch I created under my LP usermame? Keep it? Bin it and start over next time I need to propose a change to that packaging? [18:56] I rejected it [18:56] cause its a test :P [18:57] but thats pretty cool, I can add comments and stuff [18:57] * acheron_uk is still dealing with the rejection [18:58] LOL you already fixed kcalcore :P [18:59] did you get an email notification? [18:59] yes [19:00] good. I got a bounce message from kubuntu-devel list, presumably as I wasn't subbed then under my LP email [19:01] preumably it also went to kubuntu-packagers as well [19:02] acheron_uk: could you write a guide to how you do that? [19:02] to put where? [19:03] hopefully the Dojo :) [19:04] ok, I really meant in what sort of format [19:04] how would that be presented in the dojo? [19:04] not sure yet [19:04] bullet point style list? [19:04] just basic instructions [19:05] * clivejo has never done this before [19:05] most of my stuff was git diffs [19:05] I'll just do it as a bullet point list then that you could have on a slide, or adapt to something else [19:06] can you propose a totally new git repo, or does it have to already exist? [19:06] I just pushed to my account and it all created itself [19:09] nice [19:09] it was ridiculously easy [19:10] well thats the main reason to move to LP [19:10] hopefully make it easier for people to help maintain the packaging [19:10] that's partly why I asked about whether I should bin it after each time, as it was so simple to create each time [19:11] yes having it all together is good [19:14] Ive pushed most of FW5.23 to LP and PPA [19:15] there are 3 Ill have to do manually [19:17] anything actually built? [19:17] yup [19:17] ECM [19:17] http://qa.kubuntu.co.uk/ppa-status/frameworks/build_status_5.23.0_yakkety.html [19:18] that makes sense [19:18] but the rest will need a poke [19:19] Once the stuff builds I will test it on my Yakkety VM and report the results. :) [19:19] I need to make another one of those [19:19] running kubuntu-retry-builds script [19:20] the YY VM I have started off as xubuntu, as I couldn't get the YY iso to install [19:23] my internet seems to be misbehaving [19:29] I assume rick has not scheduled the next dojo? [19:29] not sure [19:30] theres a podcast tomorrow, so probably find out more then [19:30] didn't know that. TY [19:31] https://www.youtube.com/channel/UC-ChyPPcJSMUw2au2UyIKwQ [19:32] set yourself a reminder :) [19:32] reminder set [19:34] * clivejo thinks kubuntu needs more how-to clips on Youtube [19:35] especially now plasma 5 has changed a lot, and many will be upgrading from trusty [19:37] this one wasn't bad https://www.kubuntuforums.net/showthread.php?70261-Excellent-Plasma-desktop-tutorial [20:02] hiho [20:04] hi soee_ [20:34] wee My S5 Neo is installing Android 6 update... finally [20:40] This (https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/1577540) is still affecting me. [20:40] Launchpad bug 1577540 in ubiquity (Ubuntu Yakkety) "ubi-console-setup failing on yakkety images" [Undecided,Confirmed] [20:41] If it is widespread, do not be so quick to delete your YY partitions / VMs because it might be difficult to get them back [20:57] testing installing from the daily ISO in a VM now [20:58] marco-parillo: http://i.imgur.com/7S6K9wB.png [21:05] yofel: Theres a missing symbol here worrying me, can you can a look? https://launchpad.net/~kubuntu-ppa/+archive/ubuntu/staging-frameworks/+build/9881240 [21:10] acheron_uk: That is it. I will add to the bug report. [21:12] marco-parillo: and if I try to ignore and continue, eventually get http://i.imgur.com/W2Rp5FU.png [21:12] and that is on the 2nd go, after one failure the same already [21:13] If it is pretty unviversal, I wonder why this is not marked as a critical bug. [21:13] just going to try xubuntu daily to see what happens to that [21:18] hey lovely people [21:18] marco-parillo: same on xubuntu [21:19] hi jimarvan :) [21:19] hi jimarvan [21:20] TY [21:20] how are you guys? [21:20] god I am exhausted [21:20] too hot [21:20] and really realy fed up with LP [21:22] :D [21:35] gn peeps, see you tomorrow! :) [22:09] acheron_uk: if you have patches for FW5.23 feel free to submit them [22:25] clivejo: I'll have a look in the morning [22:26] no prob