14:02:10 <mclasen> #startmeeting Atomic Workstation SIG
14:02:10 <zodbot> Meeting started Mon Mar 19 14:02:10 2018 UTC.  The chair is mclasen. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:02:10 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
14:02:10 <zodbot> The meeting name has been set to 'atomic_workstation_sig'
14:02:27 <dustymabe> .hello2
14:02:27 <zodbot> dustymabe: dustymabe 'Dusty Mabe' <dusty@dustymabe.com>
14:02:30 <kalev> .hello2
14:02:32 <zodbot> kalev: kalev 'Kalev Lember' <klember@redhat.com>
14:02:37 <otaylor> dustymabe: I'm not sure if there's acl's involved for the person starting the meeting...
14:02:41 <mclasen> ok, we only have owen for 30 minutes, and sanja can't make it today, so lets start quickly
14:02:49 <otaylor> .hello otaylor
14:02:50 <zodbot> otaylor: otaylor 'Owen Taylor' <otaylor@redhat.com>
14:03:19 <otaylor> #topic Agenda
14:03:34 <mclasen> ok, here is what is in pagure: https://pagure.io/fedora-atomic-workstation/issues
14:03:44 * otaylor experiments with the ACL system ... OK will let mclasen chair
14:04:01 <otaylor> mclasen: Suggest we discuss the meeting time first
14:04:08 <mclasen> good idea
14:04:20 <mclasen> any other agenda items to add ?
14:04:29 * dustymabe lurks :)
14:04:34 <mclasen> I have a few informational things
14:04:57 <mclasen> so lets start with
14:05:00 <mclasen> #topic  meeting time
14:05:35 <otaylor> I'm proposing that we match Workstation working group meeting and use 9am eastern rather than fixed to UTC - that means an hour earlier than today for those in US/Eastern, and this time for those in Europe
14:06:06 <mclasen> does fixing it to utc have any practical advantages ?
14:06:35 <dustymabe> some contries don't observe UTC
14:06:36 <mclasen> it seems less convenient for any participant who's in a dst-affected locale
14:06:46 <dustymabe> but a lot do
14:06:54 * kalev doesn't see any advantages to fixing it to utc
14:06:57 <otaylor> I think it happened by accident, and because I quoted a time in UTC initially
14:07:13 <dustymabe> we fix it to UTC for the atomic WG meeting
14:07:13 <aday> mclasen: people in europe might not be aware of when US DST changes, i suppose
14:07:30 <mclasen> true, we'll have 2 weeks of confusion either way
14:07:38 <dustymabe> yep
14:07:39 <mclasen> thats just the nature of dst
14:07:42 <aday> utc implies that people know time changes for their location and adjust accordingly
14:08:05 <aday> although, i'm personally fine to followw EST
14:08:40 <mclasen> anybody against following EST ?
14:09:12 <mclasen> we can also follow a European timezone, but that changes who gets to deal with the 2 week window...
14:09:55 <otaylor> Sounds like we should follow US/Eastern and move on
14:10:08 <mclasen> ok
14:10:18 <dustymabe> probably worth sending this semi-decision to the list
14:10:23 <mclasen> #agreed make the meeting time follow us/eastern
14:10:24 <dustymabe> and at least allowing for more input.
14:10:37 <mclasen> #action announce the meeting time 'change' on the list
14:10:40 <dustymabe> but I'm generally +1
14:11:32 <mclasen> ok, going through the pagure issues next
14:11:53 <mclasen> #topic https://pagure.io/fedora-atomic-workstation/issue/4
14:11:58 <mclasen> rawhide installer iso
14:12:20 <mclasen> I filed that this morning, after looking at adamw's nightlies collection
14:12:23 <otaylor> It looks like the issue is that the installed system hangs and doesn't boot?
14:12:31 <mclasen> it still lists images from fall of 2017 for faw
14:12:37 <otaylor> https://openqa.fedoraproject.org/tests/206895
14:12:41 <mclasen> we don't have working, newer isos ?
14:12:51 <dustymabe> everything looks up to date to me?
14:13:21 <dustymabe> ahh
14:13:24 <otaylor> mclasen: But there's also a problem in that there's both Workstation dvd-ostree and AtomicWorkstation dvd-ostree
14:13:28 <otaylor> mclasen: in the rawhide colum
14:13:36 <otaylor> which may be what is confusing you
14:13:44 <dustymabe> so we renamed things recently
14:13:46 <mclasen> otaylor: what does that show ? we succeed installing, but then don't get to the login screen ?
14:13:48 <otaylor> that sounds like something that adamw needs to fix in his scripts
14:13:58 <mclasen> ah, ok
14:14:03 <dustymabe> from 'workstation ostree' to atomicworkstation
14:14:11 <mclasen> #action mclasen to follow up with adamw on faw iso rename
14:14:40 <dustymabe> #info link to last night's FAW dvd: https://kojipkgs.fedoraproject.org/compose/rawhide/Fedora-Rawhide-20180319.n.0/compose/AtomicWorkstation/x86_64/iso/
14:15:04 <otaylor> actually, https://openqa.fedoraproject.org/tests/207057 - the f28 failure - is a bit different in that there's text
14:15:19 <mclasen> sounds like we do produce isos at least
14:16:08 <mclasen> I'll have a chat with adamw about the current state of affairs
14:16:23 <mclasen> anything else on this topic ?
14:16:40 <otaylor> It seems like first step would be for someone to try a clean install from the f28 iso and see if it works, or reproduces the openqa failure
14:17:00 <mclasen> kalev: did you do a fresh install recently ?
14:17:05 <mclasen> for the rpm-ostree work ?
14:18:36 <mclasen> moving on to the next topic...
14:18:47 <mclasen> #topic https://pagure.io/fedora-atomic-workstation/issue/3
14:18:53 <mclasen> confusing grub menu
14:19:13 <mclasen> does not look like anything has happened on that since we last looked at it
14:19:54 <aday> jimmac and i are going to be reviewing the boot options designs with members of aruiz's team soon
14:20:36 <mclasen> makes sense to consider this issue in that context, I would say
14:21:18 <aday> i'm guessing that there isn't design guidance for this yet?
14:21:23 <otaylor> There's nothing that would make "hit i for more information" impossible to implement, other than probably would require patches to grub (and extension to the boot loader spec)
14:21:56 <mclasen> grub is possibly the worst environment to try and build more extensive ui in...
14:22:01 <kalev> mclasen: nope, I used my old install and just updated
14:22:23 <otaylor> aday: Probably good to get some sense from pjones about what is possible before spending too much time on what we would like :-)
14:22:28 <mclasen> kalev: want to kick off a fresh install in a vm in the background, just to see if it works ?
14:22:33 <kalev> mclasen: sure
14:22:55 <aday> otaylor: yep, we're going to have a call with him
14:23:03 <mclasen> #action kalev will try a fresh install with a current faw iso
14:23:57 <mclasen> anything else on boot menus ?
14:24:03 <otaylor> aday: So I think the difference here is that the grub entries don't represent "a kernel" they represent "an operating system install" and with package layering, that operating system instance is one that has been configured locally
14:24:45 <mclasen> #info aday and jimmac will do a design review of boot options, and will look at the 'unclear grub menu' issue
14:25:46 <aday> sounds like i'll need to do a bit of research on package layering
14:25:47 <mclasen> ok, lets see if we can get one more topic in before owen has to leave
14:25:57 <mclasen> #topic https://pagure.io/fedora-atomic-workstation/issue/2
14:26:02 <mclasen> unwanted polkit dialogs
14:26:38 <mclasen> I made a patch for fedora-release(-workstation), which was slow to be merged due to some back and forth on the ticket
14:26:53 <kalev> thanks mclasen for coming up with the fixes
14:26:59 <mclasen> now I'm worried that ausil is moving on and the ticket may end up being stranded
14:27:07 <otaylor> "Merged by ausil an hour ago "
14:27:15 <mclasen> oh, yay
14:27:17 <kalev> ohhh, nice
14:27:18 <mclasen> I didn't see that
14:27:23 <mclasen> also in f28 ?
14:27:51 <kalev> I think you need to do a separate PR for f28, that seems to be the process there :(
14:28:25 <mclasen> #action mclasen will create a fedora-release PR for f28
14:28:38 <mclasen> progress, at least!
14:28:49 <kalev> indeed!
14:28:57 <mclasen> last pagure issue:
14:29:02 <mclasen> #topic https://pagure.io/fedora-atomic-workstation/issue/1
14:29:07 <mclasen> rpm-ostree support in g-s
14:29:12 <mclasen> kalev, want to give a quick update ?
14:29:25 <dustymabe> otaylor: before you leave make sure to #chair other people
14:29:28 <kalev> I'm working on fixing that :)
14:29:41 <otaylor> dustymabe: mclasen is the meeting owner this week
14:30:14 <kalev> pretty much just started looking at it and don't have much progress yet
14:30:56 <mclasen> I have a question around the UX here
14:31:14 * otaylor waves
14:31:19 <otaylor> later everybody
14:31:23 <mclasen> with rpm-ostree, the update happens in the background, and the reboot is just as quick as always
14:31:39 <mclasen> do we change anything in the g-s offline update setup to take that into account ?
14:32:19 <aday> mclasen: show something special during boot, you mean?
14:32:29 <mclasen> or rather, don't ?
14:32:56 <mclasen> I mean, with classic offline updates, we stage the update, then boot into offline update mode, then reboot into the new system
14:33:14 <mclasen> with atomic, we deploy the update, then boot into the new image
14:33:26 <mclasen> as you can see, one less reboots, and no need for a special mode
14:33:31 <kalev> we don't use the offline update mode at all, just directly boot into the update
14:33:43 <mclasen> ah, ok
14:33:46 <kalev> the offline update mode was implemented in packagekit which is completely gone from the picture here
14:33:56 <mclasen> I see, ok
14:34:19 <mclasen> great to see a fragile piece of plumbing go away
14:34:23 <kalev> indeed
14:34:28 <aday> there probably ought to be something to communicate that you've booted into the update
14:34:42 <mclasen> yes, we may still want to show the update info thing
14:34:46 <kalev> maybe we can show a gnome-shell notification from gnome-software?
14:34:47 <aday> could be similar to the existing "update successful" notification
14:34:51 * kalev nods.
14:35:15 <kalev> I'll try to hook something up
14:35:28 <aday> 👍
14:35:30 <mclasen> I guess this will be detailed in design review, once kalev has something working
14:35:40 <aday> yes, i'll make a note
14:35:45 <mclasen> #info kalev is working on rpm-ostree support in g-s now
14:36:04 <mclasen> #info offline update mode is not used here, it is just a plain reboot
14:36:29 <mclasen> kalev, you're still hopeful to get something working for f28 ?
14:36:34 <kalev> definitely!
14:37:44 <mclasen> ok, I guess thats all for this topic for today
14:38:01 <mclasen> #topic open floor
14:38:13 <aday> i assume we might want something in the restart dialogs and so on, too. "reboot to complete update" or simalar
14:38:25 <kalev> ah yes
14:38:33 <aday> and something equivalent in software
14:38:33 <mclasen> aday: good point, we need to tell the user
14:39:36 <mclasen> I have one bit of info I wanted to share
14:40:04 <mclasen> I got the FAW iso added to osinfo-db, so it will show up in Boxes' new box wizard
14:40:24 <mclasen> that should be in Fedora very soon
14:40:47 <aday> neat
14:41:07 <mclasen> I just copied the data we have for other fedora entries in osinfo-db - if we want to differentiate there, please let me know (or send suggestions directly to the libosinfo list)
14:41:20 <mclasen> #info FAW is now in osinfo-db
14:41:43 <mclasen> dustymabe: I was wondering if atomic host should be represented there as well
14:42:33 <dustymabe> mclasen: sorry. I can't answer questiosn right now
14:42:41 <dustymabe> on a call
14:42:55 <mclasen> ok
14:43:43 <mclasen> if nothing else, lets end early
14:44:28 <mclasen> next meeting would be april 2, which may be Easter Monday and a vacation day in some places ?
14:45:06 <aday> yes it is here
14:45:42 <kalev> I'll be travelling back from Estonia and probably can't join then
14:46:08 <mclasen> ok, I'll mention that conflict in the meeting time mail, we can either cancel or figure out an alternative time on the list
14:47:00 <mclasen> #endmeeting