14:00:33 #startmeeting nova 14:00:34 Meeting started Thu Aug 10 14:00:33 2017 UTC and is due to finish in 60 minutes. The chair is mriedem. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:35 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:38 The meeting name has been set to 'nova' 14:00:41 o- 14:00:47 \o 14:00:59 o/ 14:01:05 o/ 14:01:28 o/ 14:01:31 \o 14:02:08 alright, 14:02:08 \o 14:02:14 #link agenda https://wiki.openstack.org/wiki/Meetings/Nova 14:02:24 #topic release news 14:02:31 #link Pike release schedule: https://wiki.openstack.org/wiki/Nova/Pike_Release_Schedule 14:02:36 #info August 10 (today) is Pike RC1 14:02:52 there is one change i'm tracking before we cut rc1 14:02:56 #link tracking https://review.openstack.org/#/c/491012/ for rc1 14:03:23 on it 14:03:28 check queue is at 6 hours 14:03:50 there are some other changes i'm watching too and rechecking, but that's the main one 14:04:01 i can put those all at the top of the etherpad after the meeting 14:04:13 #action to summarize which changes we're watching for rc1 in the etherpad at the top 14:04:29 #info Blueprints: 50 targeted, 50 approved, 49 completed (+3 from last week) 14:04:42 the only one not marked completed yet is the doc migration one, 14:04:52 but i think we can probably mark that complete at any point since most stuff has moved over 14:04:57 sdague: or did you have a definite end state? 14:05:48 ok sdague and bauzas are still busy i guess 14:05:57 #link Pike RC1 TODO etherpad: https://etherpad.openstack.org/p/nova-pike-release-candidate-todo 14:06:13 anyone have anything they want to mention for the release? 14:06:36 yes, the manual imports 14:07:11 we really want to do those before the pike branch, otherwise we just don't have them in pike 14:07:22 sdague: so all of https://review.openstack.org/#/q/topic:bp/doc-migration+status:open+project:openstack/nova ? 14:07:25 today? 14:07:51 they won't be in the pike branch, but the docs will be generated and published to latest from master which is what i think most people are looking at 14:08:09 except they all cross link 14:08:19 so, minimally you want to get to here - https://review.openstack.org/#/c/491853 14:08:28 * bauzas notes 14:08:41 people get sent to the install guide per release 14:08:46 but this is one docs tree now 14:08:58 so every link from the pike install guide will be to pike versions of all the rest of it 14:09:01 ok i'll star that one then 14:09:19 #link https://review.openstack.org/#/c/491853 is end goal for docs migration in pike 14:09:20 which, is it's own sort of issue called out here https://review.openstack.org/#/c/492124 14:09:51 and probably deserves a dedicated carve out at PTG to figure out the dust settling of all of this 14:10:40 anyway, we can move on 14:10:46 alright 14:10:53 #topic bugs 14:11:02 nothing marked critical 14:11:18 0 untriaged right now 14:11:23 there was one for live migration from the xen team last night 14:11:29 but now i don't see it - must have been triaged, 14:11:34 but it looked related to cells v2 14:11:39 sdague: did you happen to see ^ ? 14:11:42 or did a bot take care of that? 14:11:47 it's confirmed yeah 14:11:57 https://bugs.launchpad.net/nova/+bug/1709594 14:11:58 Launchpad bug 1709594 in OpenStack Compute (nova) "live-migration without '--block-migrate" failed with "No sql_connection parameter is established"" [High,Confirmed] 14:12:01 mriedem: I confirmed it and put it under xenserver cellsv2 14:13:08 hmm, ok, i'm just wondering if it's a pike-rc-potential 14:13:17 since i don't know what would be different about this for xenserver 14:13:34 yeh, I don't know 14:14:06 ok i'm going to tag it pike-rc-potential so we can investigate it for a possible rc2 14:14:10 oh, just saw someone forgot to put back the new state for https://bugs.launchpad.net/nova/+bug/1709236 14:14:11 Launchpad bug 1709236 in OpenStack Compute (nova) "Live migration failed in openstack on xenserver" [Undecided,Incomplete] 14:14:13 after replying 14:14:29 * bauzas likes BZ for that 14:14:42 by default it assumes you close your needinfo tag 14:15:23 tl;dr: looks v2.34 regressed xen live-migration 14:15:55 ok that was newton 14:15:56 ... 14:16:13 yeh, if this goes back to newton as an issue, it's not an rc blocking bug 14:16:33 right 14:16:39 oh nevermind then 14:16:52 I'll continue digging 14:17:13 ok marked it confirmed, 14:17:16 and added ocata and newton to it 14:17:43 #info pike-rc-potential bugs: https://bugs.launchpad.net/nova/+bugs?field.tag=pike-rc-potential 14:18:00 so we've got fixes in flight for 2 of those 14:18:14 the ironic one from ed, i'd really like dansmith to be happy with that, but he's on vacation 14:18:26 and that live migration one can be an rc2 candidate 14:19:05 edleafe: dansmith asked about an ironic ci test that uses resource classes to test your healing change - i know dtantsur has a patch for that, but has there been a clean run? 14:19:20 plus talk yesterday of disallowing people from changing a resource class on a node 14:19:24 mriedem: plus https://review.openstack.org/#/c/491012/ right? 14:19:37 bauzas: yes i expect that to be approved this morning 14:19:43 or your french night 14:19:48 mriedem: I know that dtantsur is still working on that patch 14:20:15 mriedem: I can add something on top of that one 14:20:31 ok could talk about the ironic one later 14:20:38 i haven't really been following it 14:21:01 mriedem: heh, the patch is about where it was over a week ago when the ironic people told me that operators could change resource_class on a node 14:21:28 mriedem: yeah, on its way, just wanted to make sure we don't forget it :) 14:21:48 edleafe: ok i don't know if they are changing that as a bug fix in ironic or not 14:21:50 but anyway, moving on 14:22:01 gate status is ok-ish 14:22:14 there are slow nodes from citycloud-lon1 causing job timeouts and other weirdness 14:22:21 and the check queue is way backed up, as noted 14:22:25 mriedem: they are. Don't know if it's just a doc change or what 14:22:39 #topic reminders 14:22:45 yeh, just plan for > 6 hrs to get check results back at this point 14:22:49 #link Consider signing up for the Queens PTG which is the week of September 11 in Denver, CO, USA: https://www.openstack.org/ptg 14:22:56 #link Pike PTG topics etherpad: https://etherpad.openstack.org/p/nova-ptg-queens 14:23:09 sdague: ^ there is a section in there about docs, maybe add whatever you'd like to see discussed at the ptg 14:23:18 mriedem: will do 14:23:20 i need to add something about consistent home pages for the various projects 14:23:24 which i brought up in the ML 14:23:28 and seemed to have general agreement 14:25:07 #topic stable branch status 14:25:13 #link stable/ocata: https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:stable/ocata,n,z 14:25:18 #link stable/newton: https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:stable/newton,n,z 14:25:22 oodles of reviews on stable 14:25:42 yet another thing i've been meaning to flush through 14:25:52 will probably work on some of that today too 14:26:45 #topic subteam highlights 14:26:51 no cells v2 meeting this week 14:26:53 edleafe: scheduler stuff? 14:27:03 Not much to talk about, with all the hullabaloo going on with the claims patches 14:27:06 Clarified that shared resources will be not be a thing in Pike 14:27:08 Jay had to deal with a lot of (literal) shit during the meeting 14:27:09 EOM 14:27:11 Jay agreed to document the specifics of what happens (re: placement) in each of the move/resize operations 14:27:14 The meeting ended with much love for the ChanceScheduler 14:27:58 right ok those docs aren't up yet, but we have something as a stand in 14:28:13 this https://review.openstack.org/#/c/491900/ 14:28:21 also, the caching scheduler and chance schedulers are now deprecated 14:28:40 #info chance and caching schedulers are now deprecated https://review.openstack.org/#/c/492210/ 14:28:51 ...and not a tear was shed 14:28:59 oh and 14:29:05 #info ram and disk filters are no longer enabled by default https://review.openstack.org/#/c/491854/ 14:29:35 edleafe: well wait for a bit about that :) 14:29:37 alex_xu: was there an api meeting this week? 14:29:54 Also I'm working with dtantsur to deprecate all things about the ironic scheduler filters and options 14:29:58 there was not, the meetbot was down and nobody had anything to talk about so it was called off 14:30:12 ok, only api thing i know of is https://review.openstack.org/#/c/489926/ 14:30:20 which is more docs for the new route mapping 14:30:47 holy cow https://review.openstack.org/#/c/489926/9/doc/source/_static/images/evolution-of-api.png,unified 14:31:06 we should have the classic ape -> man evolution graphic in there 14:31:32 https://thumbs.dreamstime.com/z/evolution-man-25266163.jpg 14:31:51 yeh, that's a really nice patch 14:32:02 good work gmann 14:32:18 there was no notification meeting this week 14:32:35 and the nova/cinder meeting hasn't happened yet 14:32:39 but will likely be short 14:32:47 #topic stuck reviews 14:32:55 nothing on the agenda 14:32:59 does anyone want to bring something up here? 14:33:22 #topic open discussion 14:33:28 ok only thing i wanted to mention 14:33:37 #info mriedem is on vacation next week (along with sdague, bauzas and stephenfin), dansmith is in charge of release stuff in case we need an RC2 14:33:57 i don't plan on bringing my laptop with me 14:34:17 anyone have anything else to discuss? 14:34:47 ok that's it then, thanks everyone 14:34:50 ++ 14:34:52 #endmeeting