15:03:40 <v1k0d3n> #startmeeting openstack-helm
15:03:41 <openstack> Meeting started Tue Jun  6 15:03:40 2017 UTC and is due to finish in 60 minutes.  The chair is v1k0d3n. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:03:42 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:03:45 <openstack> The meeting name has been set to 'openstack_helm'
15:03:50 <v1k0d3n> hello team
15:03:55 <lrensing> o/
15:03:56 <v1k0d3n> can we get a roll call?
15:03:58 <v1k0d3n> o/
15:04:03 * portdirect not here
15:04:05 <alraddarla> o/
15:04:05 <serverascode> o/
15:04:06 <v1k0d3n> LOL
15:04:15 <v1k0d3n> portdirect: = troublemaker
15:04:22 <jayahn> o/
15:05:18 <portdirect> oh hey lamt 0/
15:05:20 <v1k0d3n> so let's start
15:05:24 <lamt> o/
15:05:27 <v1k0d3n> https://etherpad.openstack.org/p/openstack-helm-meeting-2017-06-06
15:05:59 <v1k0d3n> first topic: Kubernetes 1.6.4 work.
15:06:29 <portdirect> suppose I can go here?
15:06:44 <v1k0d3n> so as portdirect and others have outlined, 1.6.3 should be completely avoided for OSH deployments, and we have  made really good progress towards 1.6.4+
15:06:49 <v1k0d3n> yes portdirect that would be great
15:06:50 <portdirect> thankfully there was not much that needed done :)
15:07:14 <portdirect> we are running 1.6.4 in zuul, and i think many devs are now using it
15:08:01 <portdirect> the differences between it and 1.6.2 are very insignificant for us, but it overcame the configmap issues that came in with the 1.6.3 kubelet
15:08:32 <v1k0d3n> so the last item really would be updating the docs for 1.6.4
15:08:44 <v1k0d3n> can i get someone to look at this and write this into the docs?
15:08:45 <portdirect> though while making the transition - I took the opportunity to shore up the kubeadm-aio container we are using in zuul and for some local dev
15:09:05 <v1k0d3n> this should be just a small change, so really low hanging item for anyone looking for PS's
15:09:15 <portdirect> i think most of it should be done - thought I may have missed a few version numbers in the multinode docs
15:09:33 <v1k0d3n> yes, that's what i'm referring to
15:09:53 <v1k0d3n> jayahn: do you have anyone that would be interested on your side?
15:11:11 <portdirect> a simple find/replace should be all thats needed :D
15:11:23 <jayahn> 1.6.4?
15:11:34 <jayahn> kubeadm-aio container?
15:11:35 <portdirect> yup
15:11:38 <v1k0d3n> yes, that's it. and while at it helm 2.3.1
15:11:48 <v1k0d3n> just verify the versions and document. easy change.
15:12:20 <jayahn> you mean documentation changes. :
15:12:20 <portdirect> though thats not really important (helm change) we should really be targeting 2.4 i think?
15:12:23 <jayahn> :)
15:12:27 <portdirect> jayahn: yup
15:12:34 <v1k0d3n> ok cool. we can move on.
15:13:13 <v1k0d3n> next topic: helm 2.4.x
15:14:05 <portdirect> this is larry i think
15:14:39 <lrensing> yeah - i think the thing holding 2.4.x back is that the ceph secrets are not being generated within the helm chart itself
15:15:04 <lrensing> i'm working to remove the sigil dependency from generating secrets on the host, but rather generate them within the actual deployment via a job
15:15:35 <lrensing> once i get that resolved (and i'm close) we should be able to press forward with helm 2.4
15:15:38 <portdirect> ^^ this was caused by helm hard failing (yay!) when a template was not present, so 2.4 should be ok (i think) for people using ceph atm
15:15:58 <portdirect> but lrensing's work will enable use for those not using ceph
15:16:29 <v1k0d3n> larry, is the launchpad issue updated by chance?
15:16:44 * v1k0d3n checking now...
15:16:47 <lrensing> i'll assign it to myself
15:16:53 <lrensing> i haven't done that yet
15:17:21 <v1k0d3n> that would be awesome. if you could just update with where we're at...that would really help anyone who'd keeping up with the project in LP.
15:17:24 <v1k0d3n> thanks!
15:17:27 <v1k0d3n> ok
15:17:45 <v1k0d3n> next moving on to the next project: Chart Testing
15:18:02 <v1k0d3n> portdirect: your stage to discuss the rally work you're standing up.
15:18:12 * v1k0d3n hands over the mic
15:18:24 <portdirect> so I've been working on implementing 'helm test' for deployments
15:18:50 <portdirect> for simplicity I've been driving these with rally - to provide a nice front-end for tempest
15:19:15 <portdirect> I'm gonna break out keystone from here: https://review.openstack.org/#/c/469244/
15:19:23 <portdirect> as i think its vertually ready to go
15:19:32 <portdirect> but would really apprecate eyes on this
15:19:54 <portdirect> you can see the output here: http://logs.openstack.org/44/469244/33/check/gate-openstack-helm-aio-basic-ubuntu-xenial-nv/9e5ecb6/logs/rally/
15:20:41 <jayahn> portdirect: we can definitely put our eyes on this.
15:20:42 <portdirect> and the point in the console they are being run: http://logs.openstack.org/44/469244/33/check/gate-openstack-helm-aio-basic-ubuntu-xenial-nv/9e5ecb6/console.html#_2017-06-06_14_54_31_106109
15:20:52 <portdirect> jayahn: that would be awesome - thanks man
15:21:22 <portdirect> once we have this in - I'd really like it if we did not merge more charts untill they have tests that verify their operation runnin in zuul
15:21:47 * anticw puts portdirect direct in the fridge
15:22:02 <portdirect> it also highlights spots in our nova/neutron charts that need work
15:22:48 <portdirect> anticw: NEW charts. - we still need to get the current ones working :)
15:23:10 <portdirect> and for that we'll need to be a lot faster and looser until they do ;)
15:25:16 <portdirect> v1k0d3n: move on?
15:26:41 <portdirect> Ok, so I'm doing some work on multinode checks as well
15:26:53 <portdirect> https://review.openstack.org/#/c/470465/
15:27:02 <portdirect> a few ps's to get there :)
15:27:15 <portdirect> but just needs tidied up now
15:27:22 <lrensing> i've been keeping an eye on that
15:27:24 <lrensing> nice work though
15:27:41 <jayahn> this is really nice work :)
15:27:59 <lamt> ++
15:28:06 <portdirect> and then I'll hit up infra and get us a 3 node gate, so we can properly start testing ceph/neutron ha etc without compromises
15:28:56 <lrensing> :) sounds good
15:29:20 <portdirect> will be good to get StaceyF on this as well
15:30:02 <portdirect> ideally if she has the bandwidth once merged i'd like to hand over future work on this for her to lead?
15:30:05 <v1k0d3n> yes, we caan move on now portdirect
15:30:59 <portdirect> so as v1k0d3n says lets move on :D
15:31:23 <portdirect> Artur has done some awesome work here: https://review.openstack.org/#/c/470326/
15:31:39 <portdirect> and I feel really bad for not having looked at it properly yet :(
15:31:42 <portdirect> sorry man!
15:32:18 <portdirect> but if jayahn and anticw in particular could cast their eyes over it it would be awesome!
15:32:41 <portdirect> from 1st glance it looks like a good summary of the conversations we had in Boston
15:34:10 <portdirect> so on to the last thing I have to talk about :)
15:34:28 <portdirect> lamt is now a core in OSH :D
15:34:35 <alraddarla> \o/
15:34:41 <portdirect> nice to have you on TIn
15:35:05 <lamt> portdirect Thank you.
15:35:17 <portdirect> really looking forward to having your security focused eyes on this
15:35:40 <lrensing> woot
15:35:54 <portdirect> and your reviews to date have been harsh but (I hope) fair :)
15:36:14 <lamt> I will continue to help with reviews.
15:36:24 <lamt> portdirect LOL, I will tone it down. :D
15:36:34 <portdirect> so - I've run out of steam
15:36:53 <v1k0d3n> thanks for helping with the meeting portdirect
15:36:54 <portdirect> anyone have anything they want to raise?
15:37:31 <v1k0d3n> for the next meeting...i will create the etherpad early so the team can add items as they come up.
15:37:34 <v1k0d3n> does that work?
15:38:09 <jayahn> yeap
15:38:18 <portdirect> +2
15:42:46 <v1k0d3n> cool sounds good.
15:42:55 <v1k0d3n> i'll close out the meeting if there's nothing else
15:43:20 <v1k0d3n> #stopmeeting openstack-helm
15:43:54 <serverascode> I believe it's endmeeting :)
15:43:54 <v1k0d3n> #endmeeting