13:00:39 <andreasn> #startmeeting Cockpit weekly meeting 2016-05-30
13:00:39 <zodbot> Meeting started Mon May 30 13:00:39 2016 UTC.  The chair is andreasn. Information about MeetBot at http://wiki.debian.org/MeetBot.
13:00:39 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
13:00:39 <zodbot> The meeting name has been set to 'cockpit_weekly_meeting_2016-05-30'
13:00:46 <andreasn> .hello andreasn
13:00:47 <zodbot> andreasn: andreasn 'Andreas Nilsson' <anilsson@redhat.com>
13:00:54 <stefw> .hello stefw
13:00:55 <zodbot> stefw: stefw 'Stef Walter' <stefw@redhat.com>
13:01:12 <larsu> .hello larsu
13:01:12 <zodbot> larsu: larsu 'Lars Uebernickel' <lars@uebernic.de>
13:01:25 <mvollmer> .hello mvo
13:01:26 <zodbot> mvollmer: mvo 'Marius Vollmer' <marius.vollmer@gmail.com>
13:01:41 <andreasn> #topic Agenda
13:02:12 <mvollmer> * super short docker storage setup update
13:02:22 <mvollmer> (ssdssu)
13:02:26 <dperpeet_> .hello dperpeet
13:02:27 <zodbot> dperpeet_: dperpeet 'None' <dperpeet@redhat.com>
13:02:27 <stefw> * Stabilizing the base1 package
13:02:45 <larsu> mvollmer: that's a valid dbus type signature :)
13:03:11 <larsu> * should we pursue getting a dbus api for timers into systemd (for the gsoc project)
13:03:11 <mvollmer> haha!
13:03:52 <andreasn> anything else for the agenda?
13:04:45 <andreasn> #topic docker storage setup update
13:05:03 <mvollmer> the necessary changes are coming to fedora 24
13:05:10 <mvollmer> atomic 1.10 is in updates-testing
13:05:17 <andreasn> all the deps?
13:05:25 <mvollmer> docker-storage-setup itself is still missing
13:05:40 <mvollmer> https://bodhi.fedoraproject.org/updates/atomic-1.10-2.git1d6aecf.fc24#comment-440048
13:06:15 <mvollmer> i think we should aim to merge for fedora 24 first, and deal with the rest later
13:06:39 <mvollmer> but let's not forget about it like I forgot iscsi for rhel
13:07:20 <andreasn> cool
13:07:25 <andreasn> anything else on that?
13:07:29 <github> [cockpit] petervo opened pull request #4504: kubernetes: Add tooltips to node charts (master...node-chart-tooltip) https://git.io/vr7tV
13:07:35 <mvollmer> i made some cards for next sprint: update fedora-atomic to fedora 24, and add a better mechanisms to install arbitrary extra packages
13:07:57 <mvollmer> so i start looking at teaming now
13:08:11 <mvollmer> eot
13:08:17 <andreasn> all right
13:08:26 <andreasn> #topic Stabilizing the base1 package
13:08:48 <stefw> Well just a heads up that work has continued removing unstable stuff from the base1 package
13:09:03 <stefw> it'll contain only stable API ... plus a few things that seem would break too many examples and components already out of tree
13:09:09 <stefw> such as mustache.js and term.js
13:09:32 <stefw> I need help reviewing https://github.com/cockpit-project/cockpit/pull/4477
13:09:38 <stefw> since most of the other work blocks on that change
13:09:47 * larsu has already started looking at that
13:10:04 <stefw> great!
13:10:13 <dperpeet_> oh, that slipped on my list
13:10:22 <dperpeet_> larsu, please let me know if I should test or look at anything there
13:10:42 <dperpeet_> you have so much to do already
13:10:47 <stefw> that's it on that change
13:10:58 <larsu> yeah it's quite the change to the build system - I think having more people try it out would be good
13:11:40 <mvollmer> yes, will do
13:12:03 <stefw> one good rule of thumb is to do a 'make clean'
13:12:11 <stefw> if you're building from a git checkout you've already done a lot with
13:12:12 <larsu> git clean!
13:12:36 <mvollmer> git clean -fdx!
13:12:43 <stefw> careful with downloaded images
13:12:53 <mvollmer> right
13:12:55 <larsu> mvollmer: clearly it's -dxf :P
13:13:10 <github> [cockpit] petervo pushed 2 new commits to master: https://git.io/vr7qW
13:13:10 <github> cockpit/master eaf79e9 Stef Walter: bower.json: Remove angular-patternfly which is not used anywhere...
13:13:10 <github> cockpit/master 0fed271 Stef Walter: Update Patternfly to 3.4.0...
13:13:15 <larsu> stefw: $TEST_DATA helps, but not everything looks at that
13:14:05 <andreasn> eot?
13:14:07 <dperpeet_> I think we're veering off topic now :)
13:14:17 <stefw> andreasn, yup
13:14:51 <andreasn> #topic possible dbus api for timers into systemd
13:15:00 <larsu> yeah what's your thoughts on that?
13:15:12 <harish_> i think dbus is the better approach
13:15:12 <stefw> on what?
13:15:17 <larsu> I talked to the systemd guys a bit, and it's quite the departure of the current model
13:15:37 <larsu> ah sorry. The idea is that systemd offers a dbus api to create timer (and possibly other) units
13:15:54 <larsu> but that would put it in the business of writing unit files, which it doesn't do yet
13:16:24 <larsu> everyone agrees that it's kind of broken that we have to use two "APIs" now: files for writing and dbus for reading
13:16:47 <larsu> but it'd be quite some work, probably longer that the summer of code period
13:17:11 <larsu> I'd be willing to pester them more about it if we think it's a good idea, but wanted hear your opinions first
13:17:17 <petervo> i think we decided to proceed for now with using files
13:17:18 <dperpeet_> I can imagine that there needs to be careful design on the systemd side for something like this
13:17:22 <petervo> for gsoc
13:17:31 <harish_> oh..
13:17:40 <petervo> but having a dbus API in the future would be nice
13:17:43 <dperpeet_> yes, I definitely see the dbus api out of scope for gsoc timeframe
13:18:01 <dperpeet_> ideally, the code should be set up in a way to make this a simple drop in
13:18:06 <dperpeet_> in case the dbus api changes
13:18:14 <harish_> yea lets try with files now make it work
13:18:18 <harish_> yea okay
13:18:37 <harish_> about timers,
13:18:51 <harish_> i hope to have next run and last run time shown in timer interface by this week
13:19:02 <larsu> harish_: cool :)
13:19:10 <harish_> then we can next start working on creation of timers
13:19:14 <mvollmer> larsu, would this be an API to write any kind of unit file?
13:19:25 <mvollmer> or an API for timers, and unit files are just implementation?
13:19:43 <harish_> unit files is better bcoz i have to write service files too
13:19:43 <larsu> nobody knows, this is completely new territory
13:19:52 <mvollmer> right
13:19:54 <harish_> along with timers
13:20:15 <harish_> https://github.com/cockpit-project/cockpit/pull/4503 here is the pull i made for next run and last run
13:20:20 <larsu> my main question is whether we think it makes sense to spend time on this
13:20:36 <mvollmer> i would say so
13:20:48 <mvollmer> at least keep it in mind
13:21:04 <harish_> gsoc is till august end
13:21:04 <mvollmer> i mean, keep notes what kind of API would be nice to have
13:21:16 <larsu> I'm leaning towards no as well, but man the current way is ugly
13:21:26 <larsu> but it's the way systemd works :/
13:21:45 <mvollmer> I'd say we can't block on this withing gsoc
13:21:56 <mvollmer> but we can try to make a first proposal or draft
13:22:06 <larsu> oh no, harish_ will definitely write the files, that's decided
13:22:17 <mvollmer> right
13:22:29 <larsu> yeah there's a feature request here: https://github.com/systemd/systemd/issues/3234
13:22:59 <dperpeet_> larsu, I just looked that up :)
13:23:19 <harish_> okay
13:23:22 <larsu> heh
13:23:31 <harish_> and andreasn help me with design :)
13:23:33 <dperpeet_> mvollmer, that issue might be a good place to weigh in for now
13:23:35 <andreasn> yup
13:24:05 <larsu> anyway, just wanted to throw this out there </topic>
13:24:16 <andreasn> #topic Open Floor
13:24:22 <harish_> thanks larsu for all the work! :)
13:24:31 <andreasn> I have one, when does Sprint 115 end?
13:24:49 <larsu> harish_: wasn't too much work - many systemd core devs live in my city :)
13:25:37 <harish_> ah, still thanks!
13:25:43 <dperpeet_> andreasn, next week I think
13:25:48 <dperpeet_> we started on Tuesday
13:25:58 <petervo> june 7th i think
13:26:06 <larsu> does anyone mind if I upgrade the Vagrantfile to fedora 24?
13:26:12 <andreasn> ok, good. Thanks
13:26:41 <dperpeet_> larsu, should we wait for release?
13:26:50 <larsu> dunno, that's why I'm asking
13:26:52 <dperpeet_> on the other hand, it is for hacking and dev
13:26:57 <dperpeet_> so why develop for old stuff, right?
13:26:59 <andreasn> the docker storage stuff wants Fedora 24 stuff. Does anything else depends on F24 stuff?
13:27:02 <larsu> dperpeet_: ;)
13:27:04 <dperpeet_> I'm fine with upgrading to f24
13:27:18 <petervo> andreasn, selinux i think
13:27:29 <andreasn> ah, yes
13:27:57 <dperpeet_> f23 has basic setroubleshoot support
13:28:10 <dperpeet_> not all the latest stuff, but you can look at alerts at least
13:28:47 <larsu> right, most things still work
13:31:07 <andreasn> will we reach a verdict during the meeting?
13:31:26 <dperpeet_> did we have any votes against upgrading?
13:31:30 <petervo> i think we are all ok with it
13:31:33 <andreasn> none from me
13:31:36 <dperpeet_> there you go :)
13:31:37 <larsu> nah, we'll reach a verdict by someone merging my (upcoming) PR
13:31:43 <dperpeet_> indeed
13:33:10 <andreasn> anything else for the open floor?
13:33:38 <andreasn> sounds like that's it
13:33:40 <andreasn> #endmeeting