09:02:47 <flwang1> #startmeeting magnum
09:02:48 <openstack> Meeting started Wed Sep  2 09:02:47 2020 UTC and is due to finish in 60 minutes.  The chair is flwang1. Information about MeetBot at http://wiki.debian.org/MeetBot.
09:02:49 <brtknr> flwang1: maybe just a open discussion will suffice, i dont have anything in particular on the agenda, maybe apart from 1.19 support
09:02:50 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
09:02:52 <openstack> The meeting name has been set to 'magnum'
09:03:02 <brtknr> o/
09:03:05 <flwang1> o/
09:03:06 <brtknr> why not :)
09:03:09 <flwang1> yep
09:03:25 <flwang1> i'm trying to build a hyperkube image for v1.19.0
09:03:28 <flwang1> but no luck so far
09:03:43 <flwang1> i'm still trying to figure out what i'm missing
09:04:42 <flwang1> and besides that, i'm working on several security enhancements
09:04:50 <flwang1> as you can see from the code review
09:06:28 <brtknr> as in the hyperkube image fails to build?
09:06:40 <brtknr> do you not agree with the approach of using binary?
09:07:18 <flwang1> i don't mind using binary
09:07:34 <flwang1> and i know we have to for that way
09:07:37 <flwang1> but, again
09:07:44 <flwang1> we can't break the upgrade
09:09:41 <brtknr> yep i think the upgrade needs to include a piece of code to stop old containers and start new processes
09:10:21 <flwang1> yep, technically it should be able to do that. i just want to highlight this is the baseline
09:11:03 <brtknr> fedora atomic -> coreos was only breaking because of the operating system incompatibility, dont see why this way it needs to break upgrades
09:11:37 <flwang1> yes
09:11:59 <flwang1> btw, did you try v1.18.8? when using v1.18.8, the prometheus-operator failed to start
09:12:35 <flwang1> i got this http://paste.openstack.org/show/797363/
09:12:41 <flwang1> any idea?
09:13:23 <brtknr> I will test now
09:13:44 <brtknr> Is it repeatable?
09:14:13 <flwang1> on my local, it is
09:14:19 <flwang1> it's repeatable
09:14:21 <brtknr> Have you tried bumping up prometheus operator version?
09:14:36 <flwang1> i'm using v8.2.2
09:14:48 <flwang1> bumped
09:15:36 <brtknr> looks like the latest chart is 9.3.1
09:17:02 <flwang1> ok. i will give it a try
09:17:29 <brtknr> flwang1: can I ask you to review this PR btw? https://github.com/kubernetes/cloud-provider-openstack/pull/1067
09:18:16 <brtknr> it is for a github action to release cinder csi and manila csi chart
09:18:42 <brtknr> but the controversy is around whether the releases should use latest or specific version tags
09:18:44 <flwang1> sure, it looks good. i will take another look tomorrow
09:20:47 <brtknr> i understand it is master branch but i think release shoudld have specific version tags
09:22:00 <flwang1> i think it's a similar issue like our heat container agent tag
09:24:20 <flwang1> http://paste.openstack.org/show/797364/
09:24:34 <flwang1> this timeout maybe happened earlier
09:25:03 <flwang1> i know you have done some work for the helm in magnum
09:25:07 <flwang1> any idea how to debug this?
09:31:35 <flwang1> #endmeeting