21:00:08 #startmeeting nova 21:00:09 Meeting started Thu Jan 4 21:00:08 2018 UTC and is due to finish in 60 minutes. The chair is mriedem. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:00:10 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:00:12 The meeting name has been set to 'nova' 21:00:17 \o 21:00:17 hi 21:00:22 o/ 21:00:55 takashin: i hope you sleep in on other days 21:01:09 o/ 21:01:16 mriedem: Hi. 21:01:20 \o 21:01:52 alright let's get started 21:01:55 #link https://wiki.openstack.org/wiki/Meetings/Nova#Agenda_for_next_meeting 21:01:59 #topic release news 21:02:05 #link Queens release schedule: https://wiki.openstack.org/wiki/Nova/Queens_Release_Schedule 21:02:11 #link Release schedule recap email: http://lists.openstack.org/pipermail/openstack-dev/2018-January/125953.html 21:02:33 thanks to efried for writing up details on the placement work 21:02:41 #info Jan 18: non-client library (oslo, os-vif, os-brick, os-win, os-traits, etc) release freeze 21:03:07 as i said in the recap email, i'll be out the week of jan 15, so i'll plan on doing the final non-client library releases at the end of next week 21:03:14 so let me know if you need to get something into os-vif or os-traits 21:03:29 #info Jan 25: q-3 milestone, Feature Freeze, final python-novaclient release, requirements freeze, Soft String Freeze 21:03:43 jaypipes Was there os-traits work you were going to do for ironic? RAID stuff or whatever? 21:03:54 maybe you did and I just wasn't watching. 21:04:06 (From the halloween meeting) 21:04:13 https://review.openstack.org/#/q/project:openstack/os-traits+status:open 21:04:21 https://review.openstack.org/#/q/project:openstack/os-vif+status:open+branch:master 21:04:28 efried: I have not done anything, no. my ironic work is focusing around aggregate affinity stuff right now. 21:04:46 ight 21:05:33 #info Blueprints: 53 targeted, 53 approved, 17 complete (2 completed since the last meeting) 21:05:47 slowly making progress 21:06:41 questions about the release? 21:06:50 #topic bugs 21:06:58 nothing marked critical 21:07:02 #info 43 new untriaged bugs (down 1 since the last meeting) 21:07:23 did some triage last night to help get those numbers down, the usual misconfiguration / support request bugs to flush out 21:07:48 #link tracking queens rc blockers here https://etherpad.openstack.org/p/nova-queens-release-candidate-todo 21:08:00 #link check queue gate status http://status.openstack.org/elastic-recheck/index.html 21:08:06 gate on master is okish, 21:08:11 gate on stable/pike has been busted for about 3 weeks, 21:08:18 there are patches floating around to get that working again 21:08:33 3rd party ci on master, i don't have any news 21:08:41 #topic reminders 21:08:44 #link queens blueprint status https://etherpad.openstack.org/p/nova-queens-blueprint-status 21:08:54 I'm continuing to update stuff in there 21:09:06 #topic stable brach status 21:09:08 #link stable/pike: https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:stable/pike,n,z 21:09:13 #link Etherpad for bugs tracked for Pike backports: https://etherpad.openstack.org/p/nova-pike-bug-fix-backports 21:09:19 #link stable/ocata: https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:stable/ocata,n,z 21:09:24 #info Ocata 15.1.0 released 21:09:30 #link stable/newton: https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:stable/newton,n,z 21:09:35 #link Newton 14.1.0 release request: https://review.openstack.org/#/c/529102/ 21:09:41 newton eol is dependent on a series of changes 21:09:48 https://review.openstack.org/#/q/I1aa3ca6cc70cef65d24dec1e7db9491c9b73f7ab 21:10:13 #topic subteam highlights 21:10:21 i see that dansmith is lurking 21:10:25 cellsv2? 21:10:37 ohai 21:10:52 we had a meeting, not a lot changed since last year being that most people weren't around 21:11:03 we went over some bugs and things that are waiting review, 21:11:20 but nothing major that is new of note that I recall 21:11:32 yup 21:11:37 edleafe: scheduler? 21:11:41 No meeting this week. 21:11:48 We're slowly emerging from the holiday coma. 21:12:10 * jaypipes working on nested resource providers integration with allocation candidates. 21:12:12 So good progress, and it should continue as we are all back 21:12:24 that's about it 21:12:34 still looking for a meeting chair? or did efried take that? 21:12:37 Some discussion about having a new host (or a new time) for sched meetings. 21:13:03 yeah, either we move the meeting or efried takes over for now 21:13:15 ok 21:13:16 at least until I can get my office situation resolved 21:13:30 edleafe What time would work? An hour later? 21:13:39 maybe 2 21:13:54 I don't get to the office now until around 9:30 CST 21:14:13 see what's available on http://eavesdrop.openstack.org/ maybe 21:14:31 Okay, well, I suppose rather than moving it and then moving it back when edleafe gets settled, I can take over for a couple weeks. 21:14:31 i know we had like a bug team meeting in the past which was dropped, so that slot should be open 21:14:38 or that 21:14:40 last time we moved it we surveyed the main contributors to find a time that worked for all 21:14:57 We don't have a lot of lead time for that right now. 21:15:08 yeah, we could do this next week 21:15:15 hold the meeting on monday as usual for now 21:15:26 ok, so how about just same time for now and settle if later once we're out of the shit show that is closing out queens 21:15:29 Right, so I'll chair for now, and if your situation doesn't "improve" long term... 21:15:32 so efried - you good for running it on monday? 21:15:36 Yes 21:15:47 cool 21:16:09 ok i'm not sure if there was an api meeting this week 21:16:14 or if anyone here was there 21:16:25 there was no notification meeting this week 21:16:32 for cinder 21:16:32 Working on multi-attach patches in nova: https://review.openstack.org/#/q/topic:bp/multi-attach-volume+status:open 21:16:41 #link Tracking multiattach issues and TODOs here: https://etherpad.openstack.org/p/multi-attach-volume-queens 21:17:05 i found some nasty qemu/libvirt issues with multiattach last night, so we've hit a bump 21:17:17 currently working on workarounds for testing 21:17:32 #topic stuck reviews 21:17:36 there was nothing in the agenda 21:17:45 #topic open discussion 21:17:50 there is one item 21:17:54 (ab2434 / hemanth-n): Blueprint review https://blueprints.launchpad.net/nova/+spec/add-physical-pci-info-for-sriov-ports 21:18:11 hi. I'd like to discuss https://blueprints.launchpad.net/nova/+spec/add-physical-pci-info-for-sriov-ports 21:18:13 my understanding is this is just adding/handling "pf_pci_slot" in the binding profile 21:18:21 correct 21:18:29 are there changes on the neutron side for this? 21:18:34 no 21:18:54 what is consuming this then? 21:19:04 vif plugin? 21:19:09 yes 21:19:50 this is the codereview https://review.openstack.org/#/c/530107/ 21:20:27 would need jaypipes or someone more familar with sriov/pci stuff to comment here 21:20:29 or moshele 21:20:47 i don't really want to add this into the mix when we're <3 weeks from feature freeze for queens 21:21:20 when can we target this then? 21:21:44 rocky is the next release 21:21:59 queens feature freeze is jan 25 21:22:15 we cut queens rc1 on feb 8 and then open master for rocky dev 21:22:27 ok 21:22:54 Unsure if this is the right venue for this, but the nova-queens-blueprint-status etherpad mentioned that Johns Hopkins needs to sync up with Sean on the certificate validation feature, but he is no longer available. Who should we be talking to? We created an etherpad (https://etherpad.openstack.org/p/queens-nova-certificate-validation) to walk through usage and testing. 21:22:55 i'm particularly uneasy about this https://review.openstack.org/#/c/530107/1/nova/pci/utils.py 21:23:08 if ^ will work for all environments 21:23:37 jackie-truong: honestly i'm not sure 21:23:52 there are lots of things in flight right now and i expect a lot of blueprints to get deferred 21:24:02 which is not the answer anyone wants to hear 21:24:29 mriedem That (utils.py method) will definitely break PowerVM. We don't have sysfs on the hypervisor. 21:24:44 We have to avoid the other sysfs 21:24:48 efried: powervm is what i was thinking of :) 21:24:49 -usin' code paths for that reason. 21:25:06 Aw, that's so sweet 21:26:08 ab2434: we already store the physical function's PCI address in the pci_devices table so there shouldn't be a need to to the get_pf_pci_by_vfpci_adddress() function at all. mriedem, since this blueprint is proposing adding (yet another) key/value to the binding profile (and nothing else), I'm ok with it. 21:26:36 jaypipes: as long as they don't try to read files from disk... 21:26:53 mriedem: there's no need for that. see my comment above. 21:26:57 mriedem: I will review the patch. 21:27:02 jaypipes: right - want to mention that in the change? 21:27:03 thanks 21:27:07 thanks 21:27:07 yes, of course 21:27:24 ok anything else? 21:27:27 mriedem: Okay. We added a link to the usage and testing etherpad in the nova-queens-blueprint-status for now. We'll keep an eye out for any questions. 21:27:46 jackie-truong: yup, thanks for that 21:28:08 ok if nothing else, we'll wrap up 21:28:23 #endmeeting