14:29:14 #startmeeting tripleo 14:29:15 Meeting started Tue Jan 10 14:29:14 2017 UTC and is due to finish in 60 minutes. The chair is EmilienM. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:29:16 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:29:18 The meeting name has been set to 'tripleo' 14:29:47 I take the opportunity to remind people that right now it's a good time to stop starting things 14:30:01 and start fixing bugs and bring more stability 14:30:23 the deadline to push a feature in TripleO is Jan 23 - Jan 27 week 14:30:33 #info deadline for features in ocata is Jan 23 - Jan 27 14:31:05 #info queen series has been created in launchpad 14:31:24 dsneddon asked for it 14:31:32 any question about release management? 14:31:56 #link https://releases.openstack.org/ocata/schedule.html 14:32:16 ok next topic 14:32:18 #topic CI 14:32:27 #link http://tripleo.org/cistatus.html 14:32:35 #link http://status-tripleoci.rhcloud.com/ 14:32:59 lot of green since last week, sounds like we're doing pretty well lately 14:33:17 we still have the undercloud-upgrade job failing, but it's under control 14:33:27 it's related to the nova/cellv2 that mwhahaha and owalsh have been working on 14:33:41 matbu: quick update on upgrade job? 14:34:09 EmilienM: we hit timeout on upgrade due to multinode 14:34:23 EmilienM: thank you slagle to poke at and help 14:34:30 what is timeouting? 14:34:36 the upgrade process? 14:35:03 EmilienM: it's related to multinode, i let slagle to be more specific 14:35:07 there were several backwards incompatible changes made to the deployed-server template 14:35:07 #action team to look http://lists.openstack.org/pipermail/openstack-dev/2017-January/109600.html and give feedback 14:35:17 #action team to look https://review.openstack.org/#/c/359060/ and https://review.openstack.org/#/c/415179/ + give feedback 14:35:19 which caused neutron ports to get recreated 14:35:43 which is exposing another bug...that haproxy.cfg is not getting the updated port IP 14:35:58 so then the upgrade hangs trying to connect to mysql during Step5 14:36:05 slagle: might be a bug in puppet 14:36:12 in puppet-tripleo eventually? 14:36:17 i'm trying to add a backwards compatible deployed-server here: https://review.openstack.org/#/c/418072/ 14:36:35 I can try to look at is today (haproxy config update), it's probably a puppet thing 14:36:58 if that doesn't work, we might just need to revert https://review.openstack.org/#/c/409523/ 14:37:01 #action EmilienM to help debug upgrade job timeout wrt haproxy config update 14:37:18 slagle: ack, thx for your help here 14:37:38 EmilienM: +1 14:37:46 I have an update to give about CI 14:37:53 EmilienM: you can look at my comment on PS 67 here: https://review.openstack.org/#/c/404831/ to see the details of the IP change 14:37:58 the good news from last week, is that the workflow is working on tripleo-quickstart 14:38:02 I've been working on a new experimental job, that aims to test composable HA architecture: https://review.openstack.org/#/c/418099/ 14:38:12 slagle: ack, I'll start from there 14:39:00 so it might be good to set push as a periodic job for now, or a silent gate 14:39:00 do we have any other useful update about CI this week? 14:39:15 matbu: for upgrade? 14:39:18 s/push/it/ 14:39:21 EmilienM: yes 14:39:31 matbu: well, we need to solve this timeout issue first 14:39:42 jobs stay as experimental until they actually work 14:39:54 I don't see any value to move it into periodic pipeline if it timeouts now 14:40:05 but indeed, when it will work, +1 for periodic job. 14:40:06 EmilienM: i mean the tripleo-quickstart upgrade job is working and doesn't timeout 14:40:12 ahhhh 14:40:18 #needmoretea 14:40:38 matbu: yes, go ahead. 14:40:52 #action matbu to move the tripleo-quickstart upgrade job into periodic pipeline 14:41:08 EmilienM: maybe it a good candidate for the transition 14:41:16 i let the CI squad decide about that 14:41:21 weshay ^^ 14:41:40 matbu, you can talk to me if you need any help with it 14:41:59 and i'll say 's/matbu/ci-squad/' to move the tripleo-quickstart ... :) 14:42:24 #undo 14:42:25 Removing item from minutes: #action matbu to move the tripleo-quickstart upgrade job into periodic pipeline 14:42:31 matbu, ya.. getting upgrade in the ovb oooq experimental is a good idea 14:42:34 #action ci squad to move the tripleo-quickstart upgrade job into periodic pipeline 14:42:34 imho 14:42:48 matbu: we're talking about periodic, not experimental 14:42:56 ah.. +1 on either 14:42:57 ya periodic would be better 14:43:06 sounds like there is no pushbash 14:43:10 pushback even 14:43:13 go4it! 14:43:18 hehe 14:43:26 anything else about CI this week? 14:43:50 #topic Specs 14:44:04 not sure we have specs to discuss this week but feel free to go ahead 14:44:16 please note that from now specs are for Pike cycle only. 14:44:35 if you've had a spec for Ocata, please update it to be in Pike folder 14:45:30 I'll open the disucssion 14:45:37 #topic open discussion 14:45:58 if we have any other topic to discuss or if there is any question, it's the right time 14:46:19 just an FYI, i've got reviews out for https://blueprints.launchpad.net/tripleo/+spec/capture-environment-status-and-logs. see https://review.openstack.org/#/c/405001/ and https://review.openstack.org/#/c/405002/ 14:46:22 feedback is welcome 14:46:25 I would gently remind people we have an etherpad for Atlanta: https://etherpad.openstack.org/p/tripleo-ptg-pike 14:46:27 #link https://etherpad.openstack.org/p/tripleo-ptg-pike 14:46:43 feel free to create some sessions ^ 14:47:31 #action team to review capture-environment-status-and-logs blueprint: 405001 and 405002 14:47:37 mwhahaha: nice work! 14:48:24 anything else for this week? 14:48:55 ok thanks folks 14:49:01 see you online 14:49:03 #endmeeting