15:00:42 <mgoddard> #startmeeting kolla 15:00:45 <openstack> Meeting started Wed Jul 22 15:00:42 2020 UTC and is due to finish in 60 minutes. The chair is mgoddard. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:46 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:49 <openstack> The meeting name has been set to 'kolla' 15:00:51 <mgoddard> #topic rollcall 15:00:54 <mgoddard> \o 15:01:12 <priteau> o/ 15:01:20 <osmanlicilegi> o/ 15:01:47 <headphoneJames> o/ 15:02:44 <yoctozepto> o/ 15:03:46 <bbezak> o/ 15:03:47 <mgoddard> #topic announcements 15:03:53 <mgoddard> #undo 15:03:54 <openstack> Removing item from minutes: #topic announcements 15:03:56 <mgoddard> #topic agenda 15:04:04 <mgoddard> * Roll-call 15:04:07 <mgoddard> * Announcements 15:04:09 <mgoddard> * Review action items from last meeting 15:04:11 <mgoddard> * CI status 15:04:13 <mgoddard> * Victoria release planning (kayobe) 15:04:15 <mgoddard> * Victoria release planning (kolla & kolla ansible) 15:04:17 <mgoddard> * Kolla klub and kall 15:04:19 <mgoddard> #topic announcements 15:04:25 <mgoddard> Anyone have any? 15:05:08 <mgoddard> #info 3 months until OpenStack Victoria release 15:05:10 <mgoddard> #link https://releases.openstack.org/victoria/schedule.html 15:05:31 <mgoddard> Always good to get an idea of where we are 15:05:35 <mgoddard> brb, doorbell 15:06:33 <mgoddard> back 15:06:44 <mgoddard> #topic Review action items from last meeting 15:06:48 <mgoddard> No action items 15:06:55 <mgoddard> #topic CI status 15:07:51 <mgoddard> How are we looking? 15:07:59 <mgoddard> #link https://etherpad.opendev.org/p/KollaWhiteBoard 15:08:05 <mgoddard> L24 15:08:20 <mgoddard> Claims to be green 15:08:24 <osmanlicilegi> pretty green with some yellow flowers 15:08:52 <mgoddard> #topic Victoria release planning (kayobe) 15:09:06 <mgoddard> ok kayobians, how are we doing? 15:09:32 <mgoddard> priteau, dougsz 15:09:39 <priteau> o/ 15:09:50 <dougsz> |o 15:10:15 <mgoddard> kayobe has been pretty quiet recently 15:10:26 <mgoddard> everyone tired after ussuri? 15:10:34 <jovial[m]> holiday season ;-) 15:10:57 <priteau> Customer deployment season ;-) 15:11:12 <dougsz> It just works (Tm) 15:11:21 <priteau> We have a few patches that could be merged soon, they already have a +2 15:11:23 <mgoddard> all good excuses 15:11:49 <priteau> https://review.opendev.org/#/q/status:open+project:openstack/kayobe+branch:master+label:Code-Review%252B2 15:12:44 <mgoddard> +1, let's at least get those in 15:13:16 <mgoddard> a few stable backports open too: https://review.opendev.org/#/q/status:open+project:openstack/kayobe+NOT+branch:master 15:13:42 <mgoddard> I remembered that we do have some CI issues with Kayobe 15:13:57 <mgoddard> overcloud upgrade jobs fail from time to time, out of disk 15:14:15 <mgoddard> is there anyone who would (ahem) like to investigate this? 15:14:34 <priteau> No time for CI work at the moment :( 15:14:58 <mgoddard> yeah, I've been busy too 15:15:10 <jovial[m]> Are we getting nodes with smaller disks occasionally? 15:15:16 <mgoddard> perhaps 15:15:44 <jovial[m]> I could put it on my backlog 15:16:01 <mgoddard> is there a quick fix we could apply/ 15:16:12 <mgoddard> prune old images after the upgrade? 15:16:16 <jovial[m]> don't suppose you have an example failure? 15:16:42 <jovial[m]> Interesting idea 15:16:47 <jovial[m]> any idea where it fails? 15:17:51 <mgoddard> looks like it is a bit more reliable recently: https://zuul.openstack.org/builds?job_name=kayobe-overcloud-upgrade-centos8 15:17:58 <wuchunyang> if we prune old images, How to ensure fast rollback ? 15:18:20 <jovial[m]> I think this is only for CI 15:18:56 <mgoddard> +1 15:20:20 <mgoddard> there is also this one: 15:20:23 <mgoddard> Failed to prepare to deploy: IPMI call failed: chassis bootdev pxe 15:20:58 <jovial[m]> that is just simulating a flaky BMC ;-) 15:21:39 <mgoddard> so every BMC? 15:22:04 <mgoddard> well let's keep an eye on kayobe CI failures, and keep the whiteboard updated 15:22:13 <jovial[m]> sounds good :) 15:22:21 <mgoddard> #topic Victoria release planning (kolla & kolla ansible) 15:22:37 <mgoddard> How are we doing with victoria?? 15:22:54 <mgoddard> Does anyone have a feature they'd like to discuss? 15:23:17 <wuchunyang> i am now doing automatic octavia 15:24:14 <headphoneJames> What is automatic octavia? 15:24:44 <wuchunyang> deploy octavia automatically 15:25:23 <yoctozepto> mgoddard: we are kinda doing more complete masakari 15:25:25 <wuchunyang> just enable octavia , and upload iamge, then octavia will work. 15:25:35 <yoctozepto> anyone with any masakari experience? 15:26:14 <mgoddard> wuchunyang: yes, good to see progress on that. Are there any questions you have about it? 15:26:40 <wuchunyang> yes , i am now just test on my aio ubuntu and 3 nodes centos8 15:27:08 <wuchunyang> i have no debian system 15:27:35 <mgoddard> would you be able to add a CI test for octavia? 15:28:00 <wuchunyang> maybe i can try to add it 15:29:13 <mgoddard> you could start with a test that just deploys octavia containers, then improve the test to use your feature when it merges 15:29:33 <mgoddard> we can help with CI tests if you like 15:30:39 <wuchunyang> is there a example to add this ? 15:32:05 <mgoddard> https://opendev.org/openstack/kolla-ansible/commit/f44876c406fcb7edf28e1d2f9ee894a220c362bc 15:32:32 <mgoddard> that one adds a prometheus/EFK test 15:33:09 <wuchunyang> good ps, i will try to add too 15:33:35 <headphoneJames> I am still struggling with TLS in nova-api 15:33:41 <wuchunyang> thanks mgoddard 15:34:21 <mgoddard> headphoneJames: still hitting the timeout issue? 15:34:32 <mgoddard> headphoneJames: have you tried looking at httpd timeout configs? 15:35:29 <headphoneJames> It's the random remote disconnect errors 15:35:56 <mgoddard> yeah, I thought it seemed to be cause by a timeout? 15:36:14 <mgoddard> have you been able to reproduce locally? 15:36:22 <headphoneJames> Not yet 15:36:44 <headphoneJames> Am going to spend more time today trying to do that 15:37:51 <mgoddard> you could try lowering timeouts in htpd 15:37:55 <mgoddard> httpd 15:38:30 <mgoddard> since CI can be slow, it hits issues you might not hit locally 15:39:43 <mgoddard> Anyone else have things to discuss re victoria features? 15:40:44 <mgoddard> #topic Kolla klub and kall 15:40:57 <mgoddard> We have a kolla kall tomorrow 15:41:14 <mgoddard> I think the plan is to continue looking at documnetation 15:41:24 <mgoddard> How did the klub go last week? 15:41:34 <mnasiadka> I made it before the end of the meeting! ;-) 15:43:10 <mgoddard> Is there anything we should discuss next week? 15:43:25 <mgoddard> https://docs.google.com/document/d/1EwQs2GXF-EvJZamEx9vQAOSDB5tCjsDCJyHQN5_4_Sw/edit# 15:44:41 <mgoddard> #topic Open discussion 15:46:53 <mgoddard> Seems people are quiet today 15:46:56 <mgoddard> Let's end the meeting 15:47:03 <mgoddard> Thanks all 15:47:06 <mgoddard> #endmeeting