21:00:18 #startmeeting nova 21:00:19 Meeting started Thu May 23 21:00:18 2019 UTC and is due to finish in 60 minutes. The chair is efried. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:00:20 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:00:22 The meeting name has been set to 'nova' 21:00:26 . 21:00:43 #link agenda https://wiki.openstack.org/wiki/Meetings/Nova#Agenda_for_next_meeting 21:00:49 \o 21:01:00 o/ 21:02:41 small party. 21:02:50 at least the cool kids are here. 21:02:52 and me 21:02:53 more beer for us! 21:03:17 okay, let's go 21:03:18 #topic Last meeting 21:03:18 #link Minutes from last meeting: http://eavesdrop.openstack.org/meetings/nova/2019/nova.2019-05-16-14.00.html 21:03:30 #link merged cycle themes this morning http://specs.openstack.org/openstack/nova-specs/priorities/train-priorities.html 21:03:30 as promised last meeting 21:03:44 fup bug 1827083 21:03:45 bug 1827083 in OpenStack-Gate "ERROR: Could not install packages due to an EnvironmentError: HTTPSConnectionPool(host='git.openstack.org', port=443): Max retries exceeded with url: /cgit/openstack/requirements/plain/upper-constraints.txt (Caused by NewConnectionError(': Failed to establish a new connection: [Errno -3] Temporary failure in name resolution',)) in vexxhost-sjc1" [Undecided,Confirmed] https://launchpad.net/bugs/1827083 21:03:45 https://review.opendev.org/#/c/659181/ is merged - does that mean the bug can be closed? 21:03:45 patch 659181 - opendev/base-jobs - Only use ipv6 dns if nodepool uses ipv6 (MERGED) - 1 patch set 21:03:51 mriedem: ^ 21:03:58 it's worked around, 21:04:04 i don't think we need to continue talking about it in our nova meeting 21:04:20 okay. 21:04:36 Review requested for a couple of patches (artom): 21:04:36 https://review.opendev.org/#/c/643023/ 21:04:36 https://review.opendev.org/#/c/643024/ 21:04:36 Need second +2/+W 21:04:37 patch 643023 - nova - Move get_pci_mapping_for_migration to MigrationCon... - 5 patch sets 21:04:38 patch 643024 - nova - Allow driver to properly unplug VIFs on destinatio... - 6 patch sets 21:04:52 These were requested a couple weeks ago. Progress has been made. 21:05:02 any other old business? 21:05:17 #topic Release News 21:06:15 I think the next major thing we're tracking is spec freeze at t-2, July 25th. 21:06:24 Seems like a long way off right now... 21:06:25 o/ 21:06:33 anything else release-y? 21:06:40 efried: we are doing spec freeze at m1 then 21:07:06 um, isn't that m2? 21:07:11 yes it's 2 21:07:14 https://wiki.openstack.org/wiki/Nova/Train_Release_Schedule 21:07:41 ah ok sorry yes m1 is june 6th 21:07:41 I wouldn't be opposed to doing a "spec proposal freeze" at m1 like keystone 21:08:13 but not married to the idea. 21:08:20 #topic Bugs (stuck/critical) 21:08:30 No Critical bugs 21:08:30 #link 85 new untriaged bugs (up 7! since the last meeting): https://bugs.launchpad.net/nova/+bugs?search=Search&field.status=New 21:08:30 #link 11 untagged untriaged bugs (up 1 since the last meeting): https://bugs.launchpad.net/nova/+bugs?field.tag=-*&field.status%3Alist=NEW 21:08:59 Gate status 21:08:59 #link check queue gate status http://status.openstack.org/elastic-recheck/index.html 21:09:20 I said last week that the gate was pretty healthy, thereby cursing it. 21:09:29 Not seeing a high percentage of spurious failures, but ye gods is it slow. 21:09:42 there was a problem yesterday 21:09:42 there was a reset yesterday 21:09:49 hopefully that will help 21:10:35 okay. I didn't notice it being any faster today. But maybe it's catching up on the backlog 21:10:57 #link 3rd party CI status http://ciwatch.mmedvede.net/project?project=nova&time=7+days 21:11:00 it's hours in queue before going to a node these days 21:11:26 yeah 21:11:51 could it be ipv6 dns workaround? :) 21:12:53 nova-multi-cell looks fairly ill for the past two or three days. It's non-voting, so maybe we don't care. 21:13:10 anything else for bugs or CI? 21:13:23 #topic Reminders 21:13:31 I got nothin. Anyone? 21:13:52 #topic Stable branch status 21:13:52 #link Stein regressions: http://lists.openstack.org/pipermail/openstack-discuss/2019-April/005637.html 21:13:52 bug 1824435 still pending 21:13:54 bug 1824435 in OpenStack Compute (nova) stein "fill_virtual_interface_list migration fails on second attempt" [High,Triaged] https://launchpad.net/bugs/1824435 21:14:12 this is marked as high, not sure it really is. 21:14:21 #link stable/stein: https://review.openstack.org/#/q/status:open+(project:openstack/os-vif+OR+project:openstack/python-novaclient+OR+project:openstack/nova)+branch:stable/stein 21:14:21 #link stable/rocky: https://review.openstack.org/#/q/status:open+(project:openstack/os-vif+OR+project:openstack/python-novaclient+OR+project:openstack/nova)+branch:stable/rocky 21:14:21 #link stable/queens: https://review.openstack.org/#/q/status:open+(project:openstack/os-vif+OR+project:openstack/python-novaclient+OR+project:openstack/nova)+branch:stable/queens 21:14:35 dropped that bug to medium 21:14:43 imacdonnell comments on https://review.opendev.org/#/c/653065/ 21:14:44 I've dropped pike off of the list ^ since it's -em now. There was one -W patch in there 21:14:44 patch 653065 - nova - Do not create default security group during instan... - 2 patch sets 21:14:48 but i haven't had time to dig into it 21:14:54 thanks mriedem 21:14:56 regarding stein, 21:15:04 i'm going to propose a release once the approved changes here https://review.opendev.org/#/q/status:open+project:openstack/nova+branch:stable/stein+label:Code-Review=2 are merged 21:15:28 ight. 21:16:00 #topic Sub/related team Highlights 21:16:09 Placement (cdent) 21:16:15 #link latest pupdate http://lists.openstack.org/pipermail/openstack-discuss/2019-May/006388.html 21:16:26 From last time: 21:16:28 #link WIP spec for nested magic https://review.opendev.org/#/c/658510/ 21:16:29 patch 658510 - placement - WIP: Spec for nested magic - 2 patch sets 21:16:41 I'm working on a new rev for ^ right now, should have it up in the next hour or so. 21:16:51 the theme will be simplification 21:16:57 #link spec for rg/rp mapping https://review.opendev.org/#/c/657582/ 21:16:58 patch 657582 - placement - Resource provider - request group mapping in alloc... - 1 patch set 21:17:06 Needs an owner 21:17:28 edleafe or anyone have further nova-related placement info to impart? 21:17:51 API (gmann) 21:17:54 Not I 21:18:10 nothing in the agenda, no gmann in channel, anyone have anything here? 21:18:22 (Discussions about API changes coming up later) 21:18:27 i, 21:18:36 will probably be proposing a simple spec to filter hypervisors by compute service host, 21:18:55 watcher has to jump through a few hoops to get from compute service host to hypervisor which could be optimized with a filter parameter in the api 21:19:26 ack 21:19:39 #topic Stuck Reviews 21:19:49 #link host/node validation in the API on server create http://lists.openstack.org/pipermail/openstack-discuss/2019-May/006518.html 21:20:03 last I saw on the ML, consensus was "do it early" 21:20:19 right? 21:20:23 yeah i'm not sure there is much to talk about here, seems everyone wants option 3 21:20:28 validate host via hostmapping, 21:20:30 validate node via placement 21:20:39 i like ^ 21:20:40 optimize later with request spec destination node_uuid magic 21:20:49 ++ 21:21:06 #topic Review status page 21:21:18 #link nova review status http://status.openstack.org/reviews/#nova 21:21:25 no progress on this from me since last week, sorry. 21:21:49 I think we asked fungi to respin the data, and he did, though it doesn't look like it has changed substantially. 21:22:07 We still don't know how the list is generated or how the fields are calculated. 21:22:20 I'll get around to it eventually. Meantime, if anyone else wants to hack, the code is at openstack/reviewday. 21:22:30 #topic Open discussion 21:22:38 #link WIP TC Vision Reflection https://review.opendev.org/658932 21:22:39 patch 658932 - nova - WIP: TC Vision Reflection - 3 patch sets 21:22:40 yeah, something had broken the installed dependencies for reviewday on the status host 21:22:42 No movement on this since last week. 21:22:43 efried: the hover text shows how teh score was dervied 21:22:53 that's fixed so it should be refreshing roughly every 10 minutes again 21:23:11 sean-k-mooney: oh dang, nice spot. 21:23:16 thanks fungi 21:24:01 asking for possibly more reviews on https://review.opendev.org/#/c/658903/ (Directly download and upload images to RBD), given the spec freeze soon 21:24:02 patch 658903 - nova-specs - Directly download and upload images to RBD - 1 patch set 21:24:38 oh yeah, that's on my list. thanks for reminding 21:24:53 ack, added self 21:25:06 (stephenfin) Approval for specless blueprint to remove nova-consoleauth https://blueprints.launchpad.net/nova/+spec/remove-consoleauth 21:25:25 existing review pointed out that a lot of "helper" functions already exists in code for boot-from-volume stuff, making implementation even easier than before (potentially). Thanks! 21:25:37 stephenfin said he wouldn't be around to defend himself, but we should tag him with any actions 21:25:41 so like all of them 21:25:54 sounds fair 21:25:55 There was a brief discussion about this today in -nova, /me finds... 21:26:12 http://eavesdrop.openstack.org/irclogs/%23openstack-nova/%23openstack-nova.2019-05-23.log.html#t2019-05-23T16:34:49 21:26:40 was any conclusion reached? Can we approve the blueprint? 21:27:18 it would be nice to have *any* anecdotal evidence that people have upgraded through to the point of not using the nova-consoleauth service at all, 21:27:26 because there was a ton of confusion over that after rocky was released 21:27:52 i know mnaser is past that point but don't know if he's killed all nova-consoleauth services 21:28:06 i guess we shoudl check with some of the deployment tools and see if they have made the switch by defualt and have an automated upgrade path 21:28:11 also no idea how far along/behind the deployment projects are in this regard 21:28:12 yeah, that is my hesitation as well. in theory it should be OK to remove it now but it's been bumpy and could use some reassurance 21:28:48 so before moving forward, i'd want to know how OSA, kolla and tripleo are regarding this first 21:28:50 okay, hypothetically what if someone is still using it? How careful do we really need to be? It's been deprecated for two releases and we've made efforts to reach out. 21:29:04 mmph, okay. 21:29:26 removing cells v1 is easy b/c we mostly knew who was using it and most tooling didn't support it, 21:29:37 nova-network is in a similar boat in that people have moved to neutron long ago, 21:29:43 I guess 21:29:43 #action stephenfin to find out how OSA, kolla, and TripleO are positioned wrt nova-consoleauth 21:29:43 which is easy for me to say. 21:29:45 but nova-consoleauth was required everywhere up until rocky 21:29:54 perhaps the ar to tag stepehen with is email the list and as the deployment prject to confrim what there current support is for deploying without it 21:30:09 just so 21:30:19 okay, I guess we move on. 21:31:09 (melwitt): Seeking approval for specless blueprint to add --before option to nova-manage db archive_deleted_rows https://blueprints.launchpad.net/nova/+spec/nova-archive-before 21:31:34 #link code for nova-archive-before https://review.opendev.org/#/c/556751/ 21:31:35 patch 556751 - nova - Add --before to nova-manage db archive_deleted_rows - 14 patch sets 21:31:38 this started out as a wishlist bug and the author went ahead and created a specless bp for it 21:32:12 being able to archive deleted rows based on time is something I think a lot of people want, I know the folks in RDO cloud really need it 21:32:14 code looks pretty ready. mriedem and dansmith have had review eyes on it. 21:32:21 huh, ok 21:32:22 yeah 21:32:27 the bp is just paperwork at this point 21:32:32 i think everyone was in line with it already anyway 21:32:36 bp seems unnecessary to me, but.. whatever 21:32:44 okay, then let's approve it and move on... 21:32:47 yeah. just wanted to bring it up here instead of ninja approving it 21:32:55 wfm 21:33:08 done 21:33:14 thanks 21:33:33 Any other opens? 21:34:21 cool, thanks all 21:34:21 o/ 21:34:21 #endmeeting