Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2024-10-29
16:06:14 tkajinam nothing I'm aware of
16:06:28 fwiesel o/
16:06:29 bauzas before moving on, I'm still awaiting the results of the periodic checks
16:06:35 bauzas gosh, it takes a while
16:07:04 tkajinam py38 was removed from master u-c so now py38 job may be completely broken in master, but we believe we already bumped the min version of all nova deliverables so no job may still run in py38 for master
16:07:13 bauzas ah, got it, all periodics in green except nova-emulation on 2023.2 stable branch
16:07:18 tkajinam https://review.opendev.org/c/openstack/requirements/+/925201
16:07:20 tkajinam just fyi
16:07:48 tkajinam next big event may be switch to noble, I expect
16:07:55 bauzas tkajinam: thanks for the note, I'll mention something about that in the next topic
16:08:11 bauzas tkajinam: when is planned that big change ?
16:08:45 tkajinam I saw gmann started submitting a few patches for it but I've seen a few projects have problems when run in noble so it may take some time
16:08:47 sean-k-mooney it should happen in the next few weeks
16:09:05 sean-k-mooney so ideally before m1 perhaps m2 at the very latest
16:09:12 tkajinam yeah
16:09:32 sean-k-mooney nova-next has been on it since last cycle
16:09:40 tkajinam https://review.opendev.org/c/openstack/nova/+/932648
16:09:41 sean-k-mooney i breifly looked at the docs failure
16:09:47 sean-k-mooney nova is in a pretty good state
16:10:00 tkajinam probably https://review.opendev.org/q/topic:%22migrate-to-noble%22 is a better link
16:10:05 sean-k-mooney i think we just need to add the right dep for pcr2 to bindep
16:10:13 bauzas yeah I'm not afraid of the noble change in the gate, but I'd like somehow to get noticed when it happens
16:10:30 sean-k-mooney well its for us to review
16:10:42 sean-k-mooney there are two parts
16:10:49 sean-k-mooney one is devstack will change its default nodeset
16:10:57 sean-k-mooney the other is we need to update our in repo jobs https://review.opendev.org/c/openstack/nova/+/932648
16:11:00 sean-k-mooney does the latter
16:11:46 frickler there's an ML thread on the noble upgrade https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/message/JOMDY26TCW7OX3NXRGOYQCIDXNNJ4E25/
16:11:51 bauzas ok, I'll cc myself on gmann's patch
16:11:56 frickler I assume gmann will update once ready
16:12:35 sean-k-mooney yep so at the momp[ent the only blocker on teh nova side that im aware of is including the correct package for prce.h
16:12:42 sean-k-mooney in the docs jobs (api ref ectra)
16:12:58 frickler ack, lots of other projects look much worse off
16:13:04 bauzas hence the failure
16:13:18 sean-k-mooney yep so in my devstack i have pcre 2 and 3 and maybe 1 installed
16:13:32 sean-k-mooney we just need to see which one actully provides the correct dep
16:13:35 bauzas ack
16:13:41 sean-k-mooney the packaging of that is a bit weired and may have changed
16:13:49 bauzas I guess we're cool then
16:13:58 sean-k-mooney more or less
16:14:02 opendevreview Merged openstack/nova master: Refactor obj_make_compatible to reduce complexity https://review.opendev.org/c/openstack/nova/+/928590
16:14:10 sean-k-mooney i think we can just review this as normal
16:14:17 sean-k-mooney and check in each week until merged
16:15:05 bauzas ok, moving on then
16:15:57 bauzas #topic Release Planning
16:16:03 bauzas #link https://releases.openstack.org/epoxy/schedule.html
16:16:09 bauzas #action bauzas to add Epoxy nova deadlines in the schedule
16:16:41 bauzas if you were not present at the PTG, you need to know that we'll have a spec soft freeze by mid-Dec
16:16:54 bauzas plan in advance to write your spec accordingly
16:17:10 bauzas the earlier the better as always
16:17:24 bauzas #topic Review priorities
16:17:30 bauzas #link https://etherpad.opendev.org/p/nova-2025.1-status
16:17:55 bauzas I moved some stuff, but I need to do a scan check of all the items in there
16:18:19 bauzas #topic PTG summary
16:18:27 bauzas for those who were not able to attend :
16:18:36 bauzas #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/HZZPADKTWNMBGRSOEMSPLN432DXZYYTL/
16:18:52 bauzas feel free to raise questions or remarks in that thread
16:19:05 bauzas #topic Stable Branches
16:19:18 bauzas hah, elodilles_pto is on PTO as the nick says
16:19:40 bauzas let's move on then to the next topic unless someone wants to raise something about stable branches
16:19:45 zigo FYI (about this link): We'd love to have "vTPM live migration" implemented... :P
16:20:01 bauzas zigo: make a wish
16:20:31 sean-k-mooney many woudl and there is at least some hope that it will be doen in epoxy
16:20:57 sean-k-mooney its non trivial however without wekening security
16:20:58 bauzas there is already some code there so the wish wouldn't be too hard to do
16:21:49 sean-k-mooney yes and no the current code is not accpabel as is as it wont work in the general case without custom policy
16:21:52 bauzas zigo: if you really like the case, please review the spec then once it's uploaded
16:22:05 sean-k-mooney but that is why we are goign to have a spec and try and delvier a better generic solution
16:22:40 bauzas zigo: as sean-k-mooney is explaining, we discussed about user bond of trust with admins
16:22:57 bauzas so I'd like your op voice to be shared on that spec
16:23:49 bauzas anyway, let's not derail the meeting into a technical discussion about a specific feature
16:24:20 bauzas moving on (and I think we can skip the stable branches topic for today)
16:24:23 bauzas #topic vmwareapi 3rd-party CI efforts Highlights
16:24:33 bauzas fwiesel: want to share something ?
16:27:47 fwiesel Yes, sorry. I found a regression in our networking driver and fixed it. But there is another one.
16:28:26 fwiesel Some race-condition in the startup. I am optimistic that I can fix it this week.
16:28:46 fwiesel That's from my side. Back to you bauzas.
16:29:03 bauzas thanks
16:29:10 bauzas and good luck with the investigation
16:29:19 bauzas #topic Open discussion
16:29:48 bauzas just refreshed the page, nothing in there
16:29:53 bauzas so, anything anyone ?
16:30:37 tkajinam I've proposed oslo.utils release so we hopefully get it soon and can use it for some work in nova https://review.opendev.org/c/openstack/releases/+/933627
16:30:38 sean-k-mooney i tought there was somethign we said to cover in the meeting instead of the ptg
16:30:40 sean-k-mooney but i dont recall
16:30:43 tkajinam dansmith, ^^^ just fyi
16:30:47 dansmith tkajinam: thanks
16:30:52 bauzas tkajinam: thanks
16:31:05 tkajinam I'll work on https://bugs.launchpad.net/nova/+bug/2083518 once the release is created
16:31:52 sean-k-mooney i might be thinkign of https://review.opendev.org/c/openstack/nova-specs/+/932653 but i highlighted that in the cidner cross project
16:32:21 tkajinam sean-k-mooney, ah, yeah
16:32:23 bauzas yeah, we had no topics that were left to be discussed
16:32:26 sean-k-mooney tkajinam: you are porting that to oslo yes
16:32:40 sean-k-mooney and we will just consume the predecate form there once its released
16:32:49 tkajinam sean-k-mooney, yes. you are correct.
16:33:09 sean-k-mooney ya so i dont see anythign contoversion with that so im happy to review that when its ready
16:33:22 tkajinam :-)
16:34:13 tkajinam sean-k-mooney, talking about the spec you raised... I think you spotted a nice point and we need a new mechanism in cinder (qos specs which are not used by cinder but by nova during attachment)
16:34:43 tkajinam I overlooked the fact that the prpoosal is trying to use the existing qos specs but these are cinder internal one and can't be used by nova
16:34:47 sean-k-mooney well frontend qos policies are use by the hypervior (libvirt in this case)

Earlier   Later