16:00:12 <gibi> #startmeeting nova
16:00:13 <openstack> Meeting started Thu Mar  4 16:00:12 2021 UTC and is due to finish in 60 minutes.  The chair is gibi. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:14 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:00:16 <openstack> The meeting name has been set to 'nova'
16:00:38 <dansmith> o/
16:00:51 <claudiub> o/
16:01:02 <gibi> o/
16:01:45 <elod> o/
16:01:47 <gibi> #topic Bugs (stuck/critical)
16:01:55 <gibi> no critical bugs
16:02:01 <gibi> #link 13 new untriaged bugs (-1 since the last meeting): #link https://bugs.launchpad.net/nova/+bugs?search=Search&field.status=New
16:02:19 <gibi> is there any specific bug we need to discuss today?
16:03:52 <gibi> #topic Gate status
16:03:57 <gibi> gate feels OK
16:04:06 <dansmith> lots of cinder fails,
16:04:20 <gibi> dansmith: is there anything we can help with those?
16:04:20 <dansmith> but we got some new quota recently apparently, so even though load is very high, we're doing a lot
16:04:37 <dansmith> gibi: I've offered, but we need some cinder people to look. I think we've got some attention on that now
16:04:51 <dansmith> meaning, I've offered to help and shared my notes
16:05:00 <gibi> thanks
16:05:31 <gibi> any nova specific failure we should look into?
16:05:38 <dansmith> not that I've seen
16:07:17 <gibi> #topic feature status
16:07:26 <gibi> I have a long list of status
16:07:33 <gibi> I will copy now
16:07:46 <gibi> #link https://blueprints.launchpad.net/nova/+spec/hyperv-rbd : merged \o/ along with the necessary os-brick lower constraint bump
16:07:51 <gibi> #link https://blueprints.launchpad.net/nova/+spec/nova-support-webvnc-with-password-anthentication : discussion about interference with TLS is still ongoing
16:07:58 <gibi> #link https://blueprints.launchpad.net/nova/+spec/compact-db-migrations-wallaby : Ocata and Pike nova_api db compaction is on the gate
16:08:03 <gibi> #link https://blueprints.launchpad.net/nova/+spec/modernize-os-hypervisors-api : the python-novaclient patch landed, the policy patch is still open
16:08:09 <gibi> #link https://blueprints.launchpad.net/nova/+spec/libvirt-default-machine-type : stephenfin needs to re-add his +2 but the series are approved
16:08:15 <gibi> #link https://blueprints.launchpad.net/nova/+spec/allow-disabling-cpu-flags : we have agreement about the config option and only a second +2 is needed
16:08:20 <gibi> #link https://blueprints.launchpad.net/nova/+spec/smarter-usb-devices : the last patch conflicts with the secure boot series
16:08:24 <gibi> #link https://blueprints.launchpad.net/nova/+spec/pci-socket-affinity : the last two patches has negative reviews
16:08:28 <gibi> #link https://blueprints.launchpad.net/nova/+spec/allow-secure-boot-for-qemu-kvm-guests : has negative reviews
16:08:31 <gibi> EOM
16:09:50 <gibi> is there any feature that is close to land but locked on a decision?
16:10:00 <gibi> s/locked/blocked/
16:11:17 * lyarwood joins late
16:11:30 <gibi> moving on
16:11:32 <gibi> #topic Release Planning
16:11:36 <gibi> Feature Freeze is at 11th of March, in 1 weeks from now
16:11:54 <gibi> We finished 5 and deferred 1 blueprints out of the 20 approved bp so far.
16:12:15 <gibi> I'm sure we will not close all the remaning 14 bps
16:12:34 <gibi> I sent a mail about the release planning to the ML
16:12:45 <gibi> as well as created a release tracking etherpad
16:12:50 <gibi> #link https://etherpad.opendev.org/p/nova-wallaby-rc-potential
16:13:53 <gibi> FFE rules are posted here #link http://lists.openstack.org/pipermail/openstack-discuss/2021-March/020772.html
16:14:35 <gibi> is there any question regarding the incoming FF and release?
16:15:46 <gibi> OK so it is clear then :)
16:15:51 * bauzas says whoops for having forgotten the meeting
16:15:55 <gibi> btw PTL and TC nomination periods are open until 9th of March: #link https://governance.openstack.org/election/ Feel free to do self nomination
16:16:17 <bauzas> not thinking about X but Y ;)
16:16:56 <gibi> :)
16:17:12 <gibi> #topic PTG planning
16:17:17 <gibi> PTG date had been announced: April 19-23, 2021
16:17:22 <gibi> Use the PTG etherpad to collect topics: #link https://etherpad.opendev.org/p/nova-xena-ptg
16:17:27 <gibi> I will create a poll to agree on the timing of the sessions
16:17:42 <bauzas> sean already wrote a long list of good discussions :p
16:18:16 <gibi> thanks for that
16:18:36 <gibi> #topic Stable Branches
16:18:45 <gibi> no outstanding issue with stable branches
16:18:51 <gibi> stable releases planned for Victoria, Ussuri, Train @ next week (after Feature Freeze?)
16:18:54 <gibi> EOM
16:19:42 <bauzas> do we need to look at some stable changes before ?
16:19:52 <bauzas> is there any urgent query for those ?
16:20:26 <elod> well, there are open patches, but I'm not aware of any urgent one
16:20:35 <gibi> me neither
16:20:38 <lyarwood> there's nothing urgent but I've got loads outstanding
16:20:49 <bauzas> ok, I can try to look at some
16:20:51 <lyarwood> I'll try to write up a list of things I'd like to ship in the same releases next week
16:21:15 <gibi> thanks
16:21:42 <elod> lyarwood: I'll try to review your list then next week :)
16:21:52 <elod> i mean the patches :)
16:21:57 <lyarwood> ack thanks
16:22:21 <gibi> #topic Sub/related team Highlights
16:22:26 <gibi> Libvirt (bauzas)
16:22:54 <bauzas> nothing to report, sir.
16:23:25 <gibi> thanks
16:23:28 <gibi> #topic Open discussion
16:23:32 <gibi> nothing on the agenda
16:23:39 <claudiub> I have a NUMA related question, if I may.
16:23:42 <gibi> claudiub: go
16:24:14 <claudiub> So, here https://docs.openstack.org/nova/latest/admin/cpu-topologies.html#customizing-instance-numa-placement-policies mentions that the NUMA nodes used for the instances are normally chosen at random
16:24:56 <claudiub> numa_fit_instance_to_host documentation mentions that it returns a new InstanceNUMATopology with its cell ids set to host cell ids of the first successful permutation or None
16:25:19 <claudiub> so, there's a mismatch there. I'm wondering which is correct.
16:25:45 <claudiub> From what I've seen, the compute nodes seem to be crammed on the NUMA node 0 with all the instances being scheduled there.
16:26:27 <claudiub> that could be problematic. Basically, I can have nodes with 50% consumed resources, but 1 NUMA node completely empty.
16:27:18 <claudiub> I've submitted a commit here too, which looks into this issue: https://review.opendev.org/c/openstack/nova/+/778740
16:27:43 <gibi> claudiub: I also think that we fill numa nodes sequentially
16:28:47 <gibi> hm, sean-k-mooney is not here today but I think they could answer this question
16:28:55 <gibi> claudiub: I suggest to bring this up on #openstack-nova
16:29:05 <claudiub> Indeed, that is what I've seen as well. But is there insight into why this was chosen? AFAIK, when live-migrating, it'll live-migrate into the same NUMA node (though I may be wrong), so this can be problematic.
16:29:57 <claudiub> will ask this in #openstack-nova then. :)
16:30:00 <gibi> thanks
16:30:15 <gibi> any other topic for today?
16:32:11 <gibi> if nothing then thanks for joining today!
16:32:17 <gibi> \o
16:32:20 <gibi> #endmeeting