16:00:34 <strigazi> #startmeeting containers
16:00:35 <openstack> Meeting started Tue Nov 14 16:00:34 2017 UTC and is due to finish in 60 minutes.  The chair is strigazi. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:36 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:00:38 <strigazi> #topic Roll Call
16:00:39 <openstack> The meeting name has been set to 'containers'
16:00:53 <strigazi> Spyros Trigazis
16:00:58 <slunkad> Sayali Lunkad
16:01:19 <mgoddard> Mark Goddard
16:01:48 <KwozyMan> Costin GamenČ›
16:01:55 <vabada> Daniel Abad
16:02:28 <strigazi> Thanks for joining the meeting slunkad mgoddard KwozyMan vabada
16:02:46 <strigazi> Today's agenda:
16:02:46 <strigazi> #link https://wiki.openstack.org/wiki/Meetings/Containers#Agenda_for_2017-11-14_1600_UTC
16:02:52 <strigazi> #topic Announcements
16:03:17 <strigazi> Magnum meeting calendar: You can import this ical file:
16:03:23 <strigazi> #link https://calendar.google.com/calendar/ical/dl8ufmpm2ahi084d038o7rgoek%40group.calendar.google.com/public/basic.ics
16:03:37 <strigazi> #topic Review Action Items
16:03:41 <strigazi> None
16:04:01 <strigazi> #topic Review Blueprints/Bugs/Ideas
16:04:12 <strigazi> Add support of k8s cluster over opensuse distro https://blueprints.launchpad.net/magnum/+spec/k8s-opensuse-support
16:04:45 <strigazi> slunkad: Do you have anything to discuss, can we start testing?
16:04:57 <slunkad> strigazi: sorry there hasn't been much work on it for some weeks because i had taken some time off
16:05:12 <slunkad> strigazi: I am working on the unit tests right now
16:06:04 <strigazi> slunkad: do you have functional tests too? These are equally important
16:06:16 <slunkad> might have to wait a little more for testing as I need to work on providing the image too
16:06:42 <slunkad> strigazi: https://review.openstack.org/#/c/507522/8/magnum/tests/functional/k8s_opensuse/test_k8s_python_client.py
16:07:17 <strigazi> if you have an image I can put it in https://fedorapeople.org/groups/magnum/ temporarily
16:07:48 <slunkad> strigazi: alright I will ping you soon with the image
16:07:54 <strigazi> slunkad: thanks
16:08:19 <strigazi> Next is: Use the heat agent and Software Deployments in heat templates
16:08:32 <strigazi> https://review.openstack.org/#/c/468816/
16:09:07 <strigazi> I'm investigating how smooth is to use ansible from the container to the host or ssh directly
16:09:36 <strigazi> There is one important operation that can't be executed from the container to the host
16:10:14 <strigazi> Creating a system container on the host requires to create a hardlink per container
16:10:40 <strigazi> and that is not possible since each container has each own namespace
16:11:14 <strigazi> That's it, otherwise the patch is ready, so please test it.
16:11:30 <strigazi> I'm trying to test in our production infra too
16:11:54 <strigazi> "it works on devstack" is not enough xD
16:12:13 <strigazi> And I have one more: Cluster upgrades
16:12:18 <strigazi> WIP: https://review.openstack.org/#/q/status:open+project:openstack/magnum+branch:master+topic:bp/cluster-upgrades
16:12:45 <strigazi> I'll udpate the spec with API examples https://review.openstack.org/#/c/433728/
16:13:14 <strigazi> Have a look if you haven't already
16:13:55 <strigazi> Oh, one last item that I need some input from mgoddard and slunkad
16:14:21 <slunkad> sure
16:14:41 <strigazi> There is this bug regading the update operation that I mentioned before
16:14:45 <strigazi> #link https://bugs.launchpad.net/magnum/+bug/1722573
16:14:45 <openstack> Launchpad bug 1722573 in Magnum "Limit the scope of cluster-update for heat drivers" [Undecided,New]
16:15:12 <strigazi> vabada from our team will work on it
16:15:18 <vabada> +1
16:15:49 <strigazi> and we are thinking if it is better to just change the existing update command or
16:16:21 <strigazi> fix the existing command, deprecate it and add a new scale command too
16:16:30 <strigazi> The reasoning behind it is
16:17:09 <strigazi> That the update commands tries to apply a patch to the cluster object which is just too generic
16:17:40 <strigazi> and at the same time we limit the command explicitly to do scale up/down the cluster
16:17:52 <strigazi> mgoddard: slunkad thoughts?
16:18:26 <slunkad> strigazi: what all is the new command expected to do?
16:18:51 <strigazi> Change the size of the worker nodes resource group
16:19:38 <slunkad> I definitely agree that the current way will not take us far especially because it limits the scope
16:20:10 <slunkad> I would be in favor of adding a new command with more scope
16:21:55 <strigazi> The existing command can work, but if we try to add more functionality to the update command we will start to have redirections for each operation.
16:22:23 <strigazi> ok then we will try some prototype implementation
16:22:56 <slunkad> right, it seems like a one time effort to have a smart method to do that
16:23:43 <strigazi> cool
16:23:53 <strigazi> let's move to open discussion
16:23:58 <strigazi> #topic Open Discussion
16:24:24 <strigazi> any other business?
16:25:27 <strigazi> said once
16:25:40 <strigazi> said twice
16:25:52 <strigazi> let's wrap it up then
16:26:01 <slunkad> bye!
16:26:09 <strigazi> Thanks for joining, see you next week :)
16:26:14 <strigazi> #endmeeting