21:13:17 <strigazi> #startmeeting containers
21:13:18 <openstack> Meeting started Tue Sep  3 21:13:17 2019 UTC and is due to finish in 60 minutes.  The chair is strigazi. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:13:19 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:13:21 <openstack> The meeting name has been set to 'containers'
21:13:27 <strigazi> #topic Roll Call
21:13:28 <strigazi> o/
21:13:31 <jakeyip> o/
21:14:20 <strigazi> #topic Stories/Taks/PatchSets
21:14:54 <strigazi> brtknr: jakeyip Regarding nodegroups
21:15:15 <strigazi> Theodoros is picking things up again,
21:15:28 <strigazi> are you ok with the dirrection if the patches?
21:15:34 <strigazi> let me list them
21:16:15 <strigazi> https://review.opendev.org/#/q/status:open+project:openstack/magnum+branch:master+topic:magnum_nodegroups
21:16:27 <strigazi> ng-{6,7,8,9}
21:16:31 <brtknr> sorry my internet is being weird
21:16:36 <brtknr> o/
21:17:05 <strigazi> They are in working state, they need some polish and unit tests
21:17:46 <strigazi> brtknr: I was asking if you are positive with the dirrection of nodegroups
21:17:50 <strigazi> https://review.opendev.org/#/q/status:open+project:openstack/magnum+branch:master+topic:magnum_nodegroups
21:17:53 <strigazi> ng-{6,7,8,9}
21:18:31 <strigazi> The functionality is alsmost all there (apart from the scaling master nodes)
21:18:46 <strigazi> which just needs some extra logic for etcd
21:19:08 <strigazi> the built of the infrastructure works
21:20:13 <jakeyip> I haven't been following nodegroup closely so it's a bit out of my depth but I'll take a look
21:22:24 <strigazi> no worries, let us know when you have a look
21:25:23 <strigazi> brtknr: you there?
21:26:41 <strigazi> jakeyip: anything you want to discuss?
21:28:17 <jakeyip> I am wondering what is state of fcos support?
21:28:54 <jakeyip> I see some discussion looks like many people working on/ interested in it
21:29:07 <strigazi> it needs more work
21:29:49 <strigazi> if we make the heat-agent work, it will be fine.
21:31:35 <flwang> sorry, i'm late
21:32:06 <jakeyip> Will it be only supported for U+ or is it possible to be back ported
21:33:21 <flwang> jakeyip: see https://review.opendev.org/678458
21:33:50 <flwang> as strigazi said, if we can get heat-container-agent work on fc30, it should be ok, otherwise, there will be quite a lot work
21:34:00 <brtknr> strigazi: yes i am happy with the direction
21:34:14 <jakeyip> Ah thanks
21:35:17 <brtknr> sorry there seems to be some kind of network outage in my area... both
21:35:54 <flwang> it's a bit depressed to see we have to leave the fedora atomic driver away :(
21:36:47 <brtknr> flwang: yes :(
21:36:49 <flwang> we have put quite a lot effort on that, and unfortunately FC30 is shifting as such an aggressive way
21:38:24 <brtknr> hopefully we can carry the excellent blueprint over
21:39:00 <flwang> btw, can you guys please review this one https://review.opendev.org/#/c/678294/ ?
21:39:10 <flwang> use fa 29 as the default image for devstack
21:39:31 <brtknr> the ignition files look quite painful to create however... why is it json instead of yaml?
21:39:56 <strigazi> I forgot to click reply of the f29 patch
21:40:12 <flwang> brtknr: you have to use fcct to convert the original 'script' to the ignition format
21:40:21 <flwang> and i agree it's really painful
21:40:27 <flwang> and not easy for debug
21:40:53 <flwang> because if there is anything wrong with your ignition file, you even can't ssh into the server to debug, that's bad
21:41:06 <brtknr> flwang: we do we have yaml version of all the fragments?
21:41:39 <brtknr> do we*
21:42:07 <flwang> strigazi: can't https://review.opendev.org/#/c/678294/5/devstack/lib/magnum@277 set the image property?
21:42:30 <flwang> brtknr: you mean for the 'old' coreos driver?
21:42:46 <strigazi> we don't need the old coreos driver
21:43:01 <strigazi> does not have anything for ignition anyway
21:43:18 <strigazi> scripts that run with the agent can be bash or python
21:43:38 <flwang> brtknr: the old coreos driver is not based on ignition
21:43:50 <flwang> strigazi: we can't use python directly, there is no python on fc30
21:44:36 <strigazi> scripts that *run with the agent* can be bash or python
21:44:58 <strigazi> they are executed in the container.
21:44:59 <flwang> strigazi: ah, yep
21:46:09 <strigazi> flwang: for nodegroups,
21:46:22 <strigazi> brtknr: I was asking if you are positive with the dirrection of nodegroups
21:46:30 <strigazi> flwang: : I was asking if you are positive with the dirrection of nodegroups
21:46:48 <strigazi> flwang: https://review.opendev.org/#/q/status:open+project:openstack/magnum+branch:master+topic:magnum_nodegroups
21:46:58 <strigazi> flwang: ng-{6,7,8,9}
21:47:33 <flwang> strigazi: i haven't got time to review the code yet, TBH
21:47:37 <brtknr> strigazi: yes, happy to hear Theodoros is going to be picking it up again
21:47:43 <flwang> i will start to review it and leave my comments on
21:48:03 <strigazi> flwang:  don't need a detailed review at the moment
21:48:09 <flwang> but overally, I'd like to see we're on the same direction with other cloud provider's NG features
21:48:23 <brtknr> i suppose we need to ensure compatibility with the upcoming transition to fcos
21:48:53 <strigazi> flwang: it is kind of a rip off of GKR
21:48:54 <strigazi> flwang: it is kind of a rip off of GKE
21:49:12 <strigazi> flwang: can have a quick look?
21:49:31 <flwang> strigazi: sure, i will
21:49:34 <strigazi> we need a +1 on the direction to put effort to something that will be blocked
21:49:49 <flwang> strigazi: no problem, i will put it on my priority list
21:49:57 <strigazi> we need a +1 on the direction to *not* put effort to something that will be blocked
21:49:58 <flwang> and i'd like to see we can get it done asap
21:50:58 <strigazi> the functionality is there apart from master ng resize/addition, it needs some work for etcd but that is a detail compared to the overall direction/changes to the datamodel of magnum
21:51:46 <strigazi> we have a prototype an scaling masters too but we wait for your input.
21:53:01 <flwang> i have interest for both ng and master resize
21:53:39 <strigazi> master resize is a detail.
21:53:40 <openstackgerrit> Feilong Wang proposed openstack/magnum master: [fedora-atomic][k8s]Disable ssh password authentication  https://review.opendev.org/677320
21:53:55 <flwang> strigazi: i understand that
21:54:32 <openstackgerrit> Feilong Wang proposed openstack/magnum master: [fedora-atomic][k8s]Disable ssh password authentication  https://review.opendev.org/677320
21:55:03 <brtknr> i wasnt sure about the extent to which the ng work was ready for review... i would have a play around with it if ttstious would ping me and let me know which parts are ready
21:55:12 <flwang> as I mentioned in the PTL nomination letter, i'd like to get the things done in next cycle, which includes  NG, master containerziation, rolling ugprade and fc30
21:59:05 <flwang> strigazi: brtknr: jakeyip: anything else we need to discuss?
22:00:20 <brtknr> hmm nothing too pressing... flwang should we change back to europe morning time?
22:00:35 <brtknr> strigazi also suggested this earlier
22:02:06 <brtknr> it seems nz is exactly 12 hours ahead of UTC...
22:03:01 <flwang> what's the Europe morning time fit you guys?
22:03:17 <flwang> i prefer 9:00 which means 9:00pm for me
22:04:22 <brtknr> flwang: that sounds good to me, if thats not too late for you
22:05:31 <flwang> i can't work too late as before :)
22:05:38 <brtknr> thats 10am at CERN
22:05:46 <flwang> but 9:00pm it's ok for me
22:05:58 <flwang> strigazi: ^
22:06:00 <brtknr> I often see you pushing commits when I am starting work in the mornings :)
22:06:04 <strigazi> 10am geneva times work perfectly
22:06:32 <flwang> brtknr: hah, yep, i love work
22:06:46 <flwang> strigazi: ok, then i will send an email to update the meeting time
22:07:00 <brtknr> jakeyip: where are you based?
22:07:08 <strigazi> 09:00 UTC is actually 11:00 Central europe
22:07:27 <strigazi> https://www.timeanddate.com/worldclock/fixedtime.html?msg=Magnum+Team+Meeting&iso=20190910T09
22:07:29 <jakeyip> I'm in Melbourne GMT+10
22:07:38 <brtknr> ah yes thats correct
22:08:19 <strigazi> in the summer I can 0900 UTC, in the winter 08OO UTC due daylight savings time
22:08:32 <jakeyip> 9 UTC is ok for me too
22:08:39 <flwang> we have daylight saving as well in nz
22:08:51 <strigazi> so sold? 0900 UTC
22:09:10 <flwang> yep, we can change it when the daylight saving start :)
22:09:22 <brtknr> great!
22:09:42 <strigazi> so as per https://www.timeanddate.com/worldclock/fixedtime.html?msg=Magnum+Team+Meeting&iso=20190910T09
22:10:36 <strigazi> end meeting?
22:10:38 <brtknr> anyone know what happened to blizzard?
22:10:38 <colin-> o/
22:10:41 <flwang> LGTM
22:10:44 <colin-> next question :)
22:10:51 <strigazi> colin-: xD
22:10:59 <brtknr> lol
22:11:00 <strigazi> no lag
22:11:04 <brtknr> bad timing
22:11:07 <colin-> hehe
22:11:19 <colin-> nothing to add this week, just saying hello
22:12:18 <brtknr> hi :)
22:12:45 <flwang> let's end the meeting
22:13:17 <strigazi> #endmeeting