16:01:59 <rdieter> #startmeeting kde-sig
16:01:59 <zodbot> Meeting started Tue Nov  8 16:01:59 2016 UTC.  The chair is rdieter. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:01:59 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
16:01:59 <zodbot> The meeting name has been set to 'kde-sig'
16:02:03 <rdieter> #topic roll call
16:02:12 <rdieter> hi all, friendly kde-sig meeting, who's present today?
16:03:22 <lupinix> .hello lupinix
16:03:22 * jreznik is here
16:03:22 <zodbot> lupinix: lupinix 'Christian Dersch' <lupinix@mailbox.org>
16:03:26 <jreznik> (hello)
16:05:17 <rdieter> #info lupinix jreznik rdieter present
16:05:21 <rdieter> #chair jreznik lupinix
16:05:21 <zodbot> Current chairs: jreznik lupinix rdieter
16:05:31 <mbriza> hello
16:05:40 * pino|work is here
16:06:22 <rdieter> #info mbriza pino|work present
16:06:25 <rdieter> #chair pino|work mbriza
16:06:25 <zodbot> Current chairs: jreznik lupinix mbriza pino|work rdieter
16:07:35 <rdieter> moving on...
16:07:38 <rdieter> #topic agenda
16:07:43 <rdieter> what to discuss today?
16:07:57 <rdieter> probably should include f25 blockers (sddm)
16:08:38 <jreznik> yep
16:09:14 <rdieter> if nothing else (and have time): fedorahosted.org retirement (ie, need to move stuff elsewhere)
16:09:36 <lupinix> nothing special here, i'll be quite quiet anyway as i have to work
16:09:42 * heliocastro here
16:09:54 <rdieter> #info heliocastro present
16:09:56 <rdieter> #chair heliocastro
16:09:56 <zodbot> Current chairs: heliocastro jreznik lupinix mbriza pino|work rdieter
16:11:07 <rdieter> ok, lets get started...
16:11:12 <rdieter> #topic f25 blockers: sddm
16:11:26 * rdieter looks up bug #
16:11:37 <mbriza> 1382001
16:11:41 <mbriza> and 1392654
16:11:50 <rdieter> .bug 1382001
16:11:50 <zodbot> rdieter: An error has occurred and has been logged. Please contact this bot's administrator for more information. - https://bugzilla.redhat.com/1382001
16:11:56 <rdieter> heh, ok
16:12:02 <rdieter> .bug 1392654
16:12:02 <zodbot> rdieter: An error has occurred and has been logged. Please contact this bot's administrator for more information. - https://bugzilla.redhat.com/1392654
16:12:03 <than> present
16:12:06 <rdieter> at leaset we have a link
16:12:10 <rdieter> #info than present
16:12:11 <rdieter> #chair than
16:12:11 <zodbot> Current chairs: heliocastro jreznik lupinix mbriza pino|work rdieter than
16:12:25 <mbriza> 1382001sddmASSIGNEDLogging out from second user after switching users kills first user's session (with xorg-x11-drv-qxl at least)
16:12:30 <mbriza> ah, tabs, sorry
16:12:58 <mbriza> 1382001 sddm ASSIGNED Logging out from second user after switching users kills first user's session (with xorg-x11-drv-qxl at least)
16:13:02 <mbriza> 1392654 sddm POST Logging out from first user after switching to second user and then back to first user crashes SDDM
16:13:42 <mbriza> while i fixed (or i'd probably say worked around) 1392654 by "manually" switching the VT
16:13:49 <mbriza> i have no idea what to do about 1382001
16:14:20 <mbriza> i've spent basically all of today debugging it and have a vague suspicion it's caused by the driver but i have nothing solid
16:15:23 <rdieter> mbriza: would testing on bare-metal help narrow if it's driver-specific or not?
16:15:35 <mbriza> it doesn't happen o bare metal at all
16:15:57 <rdieter> ok, thanks, is that clear in the bug yet?
16:16:04 <rdieter> that it is specific to qxl ?
16:16:12 <mbriza> oliver has mentioned it maybe two times but his comments are rather long
16:17:03 <mbriza> adam confirmed that too
16:17:25 <rdieter> should we push for some bug status change?  1.  re-assigned to qxl 2.  (re)consider if it's blocker-worthy  3.  some better idea?
16:18:46 <rdieter> (there's a chance qxl is still trigging a genuine sddm bug too, of course, but I guess that's hard to judge with evidence available so far)
16:19:08 <mbriza> it's really hard to debug and trace the source
16:19:14 <rdieter> mbriza: do you have advice on how best to proceed here?
16:19:18 <mbriza> maybe the driver guys would be able to tell us more about the problem
16:19:39 <mbriza> also, while testing the problem, it kept killing my (host-running) gnome boxes
16:19:59 <mbriza> could seriously be a bug in the driver that's propagated to the host too
16:20:06 <mbriza> however, no solid evidence :/
16:20:26 <pino|work> what about trying to get xorg/qxl guys involved?
16:20:36 <rdieter> if it's not a driver bug, I wouldn't expect virt/boxes to crash though
16:20:41 * Rhea thinks that someone is on wrong timezone / summer time..
16:20:43 <Rhea> o_o
16:22:21 <rdieter> ah, possible, my clocks changed this past sunday, I vaguely recall our choosing *not* to follow that in the past.  if so, my apologies
16:22:37 <rdieter> Rhea: is there another meeting scheduled for now?
16:23:30 <rdieter> https://apps.fedoraproject.org/calendar/meeting/385/?from_date=2016-11-07  looks like I'm right
16:23:33 <heliocastro> I got lost too, my goole calendar saud that would start later
16:24:03 <heliocastro> I have one single annoying bug on my system, which is lock crashing on Qt
16:24:16 <rdieter> I'll make sure to get it right next week, to be mindful of that
16:24:33 <heliocastro> And i can't identify if is Qt the guilt one or the xorg that i downgraded
16:25:15 <rdieter> I think we have a small censensus that qxl devs should get pulled in on this (to at least help debug), any comment or objection ?
16:25:31 <heliocastro> Yes, is not our problem
16:25:44 <heliocastro> We can help to debug, but far to have manpower to solve
16:25:49 <mbriza> rdieter: +1
16:26:03 <rdieter> mbriza: can you reasign the bug (or do you want me to do it)?
16:26:22 <mbriza> i'll do it
16:26:29 <rdieter> thanks.
16:26:42 <rdieter> anything else on f25-related issues/blockers?
16:27:19 <Rhea> rdieter: not this time around, but on-schedule it is, weekly 16:30 UTC
16:27:20 * rdieter will make effort to start using f25 as default setup on laptop soon
16:27:34 <rdieter> Rhea: <nod>, thx
16:27:57 <rdieter> ^^ next week we'll start an hr earlier
16:28:13 <heliocastro> I think is just we should try to go with 5.7.1 or wait for an later update ?
16:28:22 * heliocastro thinks is too  late
16:28:36 <rdieter> heliocastro: f25 is in final freeze, fyi, so... later :)
16:29:06 <rdieter> but could start working on it now (or just do rawhide initially)
16:29:20 <Kevin_Kofler> KDE SIG should normally be at 15:00 UTC, it started one hour late today indeed.
16:29:31 <heliocastro> I will do on copr first, i got some tips from buildsys guys very helpful
16:29:43 <rdieter> k, move on?
16:30:01 <heliocastro> move on
16:30:02 <rdieter> #topic open discussion
16:30:37 <rdieter> some small things from me... worked on qt-5.6.2 in copr last week, https://copr.fedorainfracloud.org/coprs/g/kdesig/qt5-lts/
16:30:59 <rdieter> and working well on my f24 box, so will probably start working committing to git and doing formal builds soon
16:31:08 <rdieter> for f24/el6
16:31:36 <rdieter> probably to skip f23 (too close to eol for any big updates)
16:31:55 <rdieter> and no el7 since qt5 is being pulled into core rhel 7.3
16:32:21 <Kevin_Kofler> I queued Calamares 2.4.4 to testing, a bunch of UEFI fixes, but there's still work needed to support shim so that grub2-efi-modules is not required on the base image (which is not present on stock Fedora live images).
16:33:06 <rdieter> I'll try to implement change to apper style patch to allow override via env var today too
16:33:23 <Kevin_Kofler> I still think you should just drop the hack entirely.
16:33:26 <Kevin_Kofler> It does more harm than good.
16:33:28 <rdieter> Kevin_Kofler: ^^ sorry for the delay
16:33:44 <Kevin_Kofler> It actually breaks rendering of the package details.
16:34:14 <rdieter> for you, unfortunately, I was not able to reproduce it :(
16:34:31 <Kevin_Kofler> Maybe it doesn't show up for updates, I haven't done updates with Apper ever since plasma-pk-updates.
16:34:35 <Kevin_Kofler> Try searching for a package by name and viewing the details below.
16:34:42 <Kevin_Kofler> (description, URL, files)
16:34:47 <rdieter> Kevin_Kofler: ah, indeed, I only checked via updates
16:34:55 <Kevin_Kofler> For me, those only render if I either click into them or switched back&forth between description and files.
16:35:23 <Kevin_Kofler> And in any case, evidence that your "fix" for rendering glitches causes… rendering glitches should be enough to bin it!
16:35:25 <rdieter> still, I'm not sure that's serious enough to warrant dropping it, imo
16:35:47 <Kevin_Kofler> It's much more serious than the harmless glitch with the progress reporting that the "fix" was added for in the first place.
16:35:57 <Kevin_Kofler> And it also makes the app look bad.
16:36:12 <Kevin_Kofler> *the "fix" also makes.
16:36:17 <rdieter> my christmas wish: the style/layout issues get fixed properly, then we'd both be happy
16:36:27 * heliocastro has a complain to Kevin_Kofler for baptize "Convert old software" as "pulling an heliocastro" :-P
16:36:42 <Kevin_Kofler> ;-)
16:37:08 <heliocastro> I should rebuild WindowMaker just as a protest
16:37:33 <rdieter> oh, another item to mention :  plans are to retire fedorahosted.org in a few months (feb 2017?), so our stuff there will need to be migrated elsewhere soon
16:37:54 <rdieter> stuff includes: fedora-kde-artwork, kde-settings
16:38:00 <heliocastro> will be EOL in favor of pagure ?
16:38:42 <rdieter> heliocastro: https://pagure.io , yes, I believe so
16:39:19 <rdieter> https://pagure.io/fedora-infrastructure/issue/5144
16:39:33 <rdieter> https://communityblog.fedoraproject.org/fedorahosted-sunset-2017-02-28/
16:40:04 <rdieter> anyone interested in starting work on migrating content ?
16:40:28 <rdieter> fwiw, fedora-kde-artwork and kde-settings repos are/were still on subversion, for extra fun
16:40:46 <Kevin_Kofler> They'll need to be converted to git for Pagure.
16:41:17 <Kevin_Kofler> Pagure continues in the usual Fedora tradition of replacing well-working Free Software with newer software that is also Free, but with only a fraction of the features.
16:42:22 * Kevin_Kofler will miss Trac.
16:44:01 <heliocastro> Well, i hope pagure has some git lfs or annex
16:44:26 <heliocastro> pass this bibaries blobs ( images ) to pagure directly will not work
16:46:01 * heliocastro need to go
16:46:14 <heliocastro> See you guys
16:47:23 <mbriza> o/
16:53:02 <rdieter> sorry, got pulled away on phone for a bit
16:53:35 <mbriza> so how will we do that, is there a support for groups on pagure or something like that?
16:53:38 <rdieter> looks like we don't have good answers right now, pagure might not be a good fit, particularly for artwork
16:53:58 <mbriza> oh yes, right on the front page, sorry for asking
16:54:00 <rdieter> i'll try to bring up the topic on list later
16:55:03 <rdieter> anything else?  (otherwise I'll close meeting in a bit)
16:55:34 <mbriza> i assume we probably should have a group on pagure anyway
16:55:45 <mbriza> regardless of where we move the repos in the end
16:56:33 <rdieter> mbriza: yes, agreed
16:58:20 <rdieter> ok, thanks everyone
16:58:23 <rdieter> #endmeeting